push app.py to huggingface

.gitignore

@@ -0,0 +1,48 @@
+.nfs*
+
+# compilation and distribution
+__pycache__
+_ext
+*.pyc
+*.pyd
+*.so
+*.dll
+*.egg-info/
+build/
+dist/
+wheels/
+
+# pytorch/python/numpy formats
+#*.pth
+*.pkl
+*.npy
+*.ts
+model_ts*.txt
+
+# onnx models
+*.onnx
+
+# ipython/jupyter notebooks
+**/.ipynb_checkpoints/
+
+# Editor temporaries
+*.swn
+*.swo
+*.swp
+*~
+
+# editor settings
+.idea
+.vscode
+_darcs
+
+# demo 
+**/node_modules
+yarn.lock
+package-lock.json
+
+# CoreML models
+*.mlpackage
+
+# Mac
+*DS_Store*

README.md

@@ -1,6 +1,6 @@
 ---
 title: EdgeSAM
-emoji: 🐢
+emoji: 🚀
 colorFrom: pink
 colorTo: yellow
 sdk: gradio

app.py

@@ -0,0 +1,493 @@
+# Code credit: [FastSAM Demo](https://huggingface.co/spaces/An-619/FastSAM).
+
+import gradio as gr
+import numpy as np
+import torch
+from segment_anything import sam_model_registry, SamAutomaticMaskGenerator, SamPredictor
+from PIL import ImageDraw
+from utils.tools_gradio import fast_process
+import copy
+import argparse
+
+parser = argparse.ArgumentParser(
+    description="Host EdgeSAM as a local web service."
+)
+parser.add_argument(
+    "--checkpoint",
+    default="weights/edge_sam_3x.pth",
+    type=str,
+    help="The path to the EdgeSAM model checkpoint."
+)
+parser.add_argument(
+    "--enable-everything-mode",
+    action="store_true",
+    help="Since EdgeSAM follows the same encoder-decoder architecture as SAM, the everything mode will infer the "
+         "decoder 32x32=1024 times, which is inefficient, thus a longer processing time is expected.",
+)
+parser.add_argument(
+    "--server-name",
+    default="0.0.0.0",
+    type=str,
+    help="The server address that this demo will be hosted on."
+)
+parser.add_argument(
+    "--port",
+    default=8080,
+    type=int,
+    help="The port that this demo will be hosted on."
+)
+args = parser.parse_args()
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+sam = sam_model_registry["edge_sam"](checkpoint=args.checkpoint, upsample_mode="bicubic")
+sam = sam.to(device=device)
+sam.eval()
+
+mask_generator = SamAutomaticMaskGenerator(sam)
+predictor = SamPredictor(sam)
+
+# Description
+title = "<center><strong><font size='8'>EdgeSAM<font></strong></center>"
+
+description_p = """ # Instructions for point mode
+
+                1. Upload an image or click one of the provided examples.
+                2. Select the point type.
+                3. Click once or multiple times on the image to indicate the object of interest.
+                4. Click Start to get the segmentation mask.
+                5. The clear button clears all the points.
+                6. The reset button resets both points and the image.
+
+              """
+
+description_b = """ # Instructions for box mode
+
+                1. Upload an image or click one of the provided examples.
+                2. Click twice on the image (diagonal points of the box).
+                3. Click Start to get the segmentation mask.
+                4. The clear button clears the box.
+                5. The reset button resets both the box and the image.
+
+              """
+
+description_e = """ # Everything mode is NOT recommended.
+
+                Since EdgeSAM follows the same encoder-decoder architecture as SAM, the everything mode will infer the decoder 32x32=1024 times, which is inefficient, thus a longer processing time is expected.
+
+              """
+
+examples = [
+    ["web_demo/assets/picture1.jpg"],
+    ["web_demo/assets/picture2.jpg"],
+    ["web_demo/assets/picture3.jpg"],
+    ["web_demo/assets/picture4.jpg"],
+]
+
+default_example = examples[0]
+
+css = "h1 { text-align: center } .about { text-align: justify; padding-left: 10%; padding-right: 10%; }"
+
+global_points = []
+global_point_label = []
+global_box = []
+global_image = None
+
+
+def reset():
+    global global_points
+    global global_point_label
+    global global_box
+    global global_image
+    global_points = []
+    global_point_label = []
+    global_box = []
+    global_image = None
+    return None, None
+
+
+def reset_all():
+    global global_points
+    global global_point_label
+    global global_box
+    global global_image
+    global_points = []
+    global_point_label = []
+    global_box = []
+    global_image = None
+    if args.enable_everything_mode:
+        return None, None, None, None, None, None
+    else:
+        return None, None, None, None
+
+
+def clear():
+    global global_points
+    global global_point_label
+    global global_box
+    global global_image
+    global_points = []
+    global_point_label = []
+    global_box = []
+    return global_image, None
+
+
+def on_image_upload(image, input_size=1024):
+    global global_points
+    global global_point_label
+    global global_box
+    global global_image
+    global_points = []
+    global_point_label = []
+    global_box = []
+
+    input_size = int(input_size)
+    w, h = image.size
+    scale = input_size / max(w, h)
+    new_w = int(w * scale)
+    new_h = int(h * scale)
+    image = image.resize((new_w, new_h))
+    global_image = copy.deepcopy(image)
+    print("Image changed")
+    nd_image = np.array(global_image)
+    predictor.set_image(nd_image)
+
+    return image, None
+
+
+def convert_box(xyxy):
+    min_x = min(xyxy[0][0], xyxy[1][0])
+    max_x = max(xyxy[0][0], xyxy[1][0])
+    min_y = min(xyxy[0][1], xyxy[1][1])
+    max_y = max(xyxy[0][1], xyxy[1][1])
+    xyxy[0][0] = min_x
+    xyxy[1][0] = max_x
+    xyxy[0][1] = min_y
+    xyxy[1][1] = max_y
+    return xyxy
+
+
+def get_points_with_draw(image, label, evt: gr.SelectData):
+    global global_points
+    global global_point_label
+    # global global_image
+
+    x, y = evt.index[0], evt.index[1]
+    # x = int(x * scale)
+    # y = int(y * scale)
+    point_radius, point_color = 10, (97, 217, 54) if label == "Positive" else (237, 34, 13)
+    global_points.append([x, y])
+    global_point_label.append(1 if label == "Positive" else 0)
+
+    print(f'global_points: {global_points}')
+    print(f'global_point_label: {global_point_label}')
+
+    draw = ImageDraw.Draw(image)
+    draw.ellipse(
+        [(x - point_radius, y - point_radius), (x + point_radius, y + point_radius)],
+        fill=point_color,
+    )
+    return image
+
+
+def get_box_with_draw(image, evt: gr.SelectData):
+    global global_box
+    # global global_image
+
+    x, y = evt.index[0], evt.index[1]
+    # x = float(x * scale)
+    # y = float(y * scale)
+    point_radius, point_color, box_outline = 5, (97, 217, 54), 5
+    box_color = (0, 255, 0)
+
+    if len(global_box) == 0:
+        global_box.append([x, y])
+    elif len(global_box) == 1:
+        global_box.append([x, y])
+    elif len(global_box) == 2:
+        global_box = [[x, y]]
+
+    print(f'global_box: {global_box}')
+
+    draw = ImageDraw.Draw(image)
+    draw.ellipse(
+        [(x - point_radius, y - point_radius), (x + point_radius, y + point_radius)],
+        fill=point_color,
+    )
+
+    if len(global_box) == 2:
+        global_box = convert_box(global_box)
+        xy = (global_box[0][0], global_box[0][1], global_box[1][0], global_box[1][1])
+        draw.rectangle(
+            xy,
+            outline=box_color,
+            width=box_outline
+        )
+
+    return image
+
+
+def segment_with_points(
+    image,
+    input_size=1024,
+    better_quality=False,
+    withContours=True,
+    use_retina=True,
+    mask_random_color=False,
+):
+    global global_points
+    global global_point_label
+
+    global_points_np = np.array(global_points)
+    global_point_label_np = np.array(global_point_label)
+
+    if global_points_np.size == 0 and global_point_label_np.size == 0:
+        print("No point selected")
+        return image, image
+
+    num_multimask_outputs = 4
+
+    masks, scores, logits = predictor.predict(
+        point_coords=global_points_np,
+        point_labels=global_point_label_np,
+        num_multimask_outputs=num_multimask_outputs,
+        use_stability_score=True
+    )
+
+    print(f'scores: {scores}')
+    area = masks.sum(axis=(1, 2))
+    print(f'area: {area}')
+
+    if num_multimask_outputs == 1:
+        annotations = masks
+    else:
+        annotations = np.expand_dims(masks[scores.argmax()], axis=0)
+
+    seg = fast_process(
+        annotations=annotations,
+        image=image,
+        device=device,
+        scale=(1024 // input_size),
+        better_quality=better_quality,
+        mask_random_color=mask_random_color,
+        bbox=None,
+        use_retina=use_retina,
+        withContours=withContours,
+    )
+
+    return image, seg
+
+
+def segment_with_box(
+    image,
+    input_size=1024,
+    better_quality=False,
+    withContours=True,
+    use_retina=True,
+    mask_random_color=False,
+):
+    global global_box
+    global_box_np = np.array(global_box)
+
+    if global_box_np.size < 4:
+        print("No box selected")
+        return image, image
+
+    masks, scores, logits = predictor.predict(
+        box=global_box_np,
+        num_multimask_outputs=1,
+    )
+    annotations = masks
+
+    seg = fast_process(
+        annotations=annotations,
+        image=image,
+        device=device,
+        scale=(1024 // input_size),
+        better_quality=better_quality,
+        mask_random_color=mask_random_color,
+        bbox=None,
+        use_retina=use_retina,
+        withContours=withContours,
+    )
+
+    return image, seg
+
+
+def segment_everything(
+    image,
+    input_size=1024,
+    better_quality=False,
+    withContours=True,
+    use_retina=True,
+    mask_random_color=True,
+):
+    nd_image = np.array(image)
+    masks = mask_generator.generate(nd_image)
+    annotations = masks
+    seg = fast_process(
+        annotations=annotations,
+        image=image,
+        device=device,
+        scale=(1024 // input_size),
+        better_quality=better_quality,
+        mask_random_color=mask_random_color,
+        bbox=None,
+        use_retina=use_retina,
+        withContours=withContours,
+    )
+
+    return seg
+
+
+cond_img_p = gr.Image(label="Input with points", type="pil")
+cond_img_b = gr.Image(label="Input with box", type="pil")
+cond_img_e = gr.Image(label="Input (everything)", type="pil")
+
+segm_img_p = gr.Image(label="Segmented Image with points", interactive=False, type="pil")
+segm_img_b = gr.Image(label="Segmented Image with box", interactive=False, type="pil")
+segm_img_e = gr.Image(label="Segmented Everything", interactive=False, type="pil")
+
+if args.enable_everything_mode:
+    all_outputs = [cond_img_p, cond_img_b, cond_img_e, segm_img_p, segm_img_b, segm_img_e]
+else:
+    all_outputs = [cond_img_p, cond_img_b, segm_img_p, segm_img_b]
+
+with gr.Blocks(css=css, title="EdgeSAM") as demo:
+
+    with gr.Row():
+        with gr.Column(scale=1):
+            # Title
+            gr.Markdown(title)
+
+    with gr.Tab("Point mode") as tab_p:
+        # Images
+        with gr.Row(variant="panel"):
+            with gr.Column(scale=1):
+                cond_img_p.render()
+
+            with gr.Column(scale=1):
+                segm_img_p.render()
+
+        # Submit & Clear
+        with gr.Row():
+            with gr.Column():
+                with gr.Row():
+                    add_or_remove = gr.Radio(
+                        ["Positive", "Negative"],
+                        value="Positive",
+                        label="Point Type"
+                    )
+
+                    with gr.Column():
+                        segment_btn_p = gr.Button(
+                            "Start", variant="primary"
+                        )
+                        clear_btn_p = gr.Button("Clear", variant="secondary")
+                        reset_btn_p = gr.Button("Reset", variant="secondary")
+
+                gr.Markdown("Try some of the examples below ⬇️")
+                gr.Examples(
+                    examples=examples,
+                    inputs=[cond_img_p],
+                    outputs=[cond_img_p, segm_img_p],
+                    examples_per_page=4,
+                    fn=on_image_upload,
+                    run_on_click=True
+                )
+
+            with gr.Column():
+                # Description
+                gr.Markdown(description_p)
+
+    with gr.Tab("Box mode") as tab_b:
+        # Images
+        with gr.Row(variant="panel"):
+            with gr.Column(scale=1):
+                cond_img_b.render()
+
+            with gr.Column(scale=1):
+                segm_img_b.render()
+
+        # Submit & Clear
+        with gr.Row():
+            with gr.Column():
+                with gr.Row():
+                    with gr.Column():
+                        segment_btn_b = gr.Button(
+                            "Start", variant="primary"
+                        )
+                        clear_btn_b = gr.Button("Clear", variant="secondary")
+                        reset_btn_b = gr.Button("Reset", variant="secondary")
+
+                gr.Markdown("Try some of the examples below ⬇️")
+                gr.Examples(
+                    examples=examples,
+                    inputs=[cond_img_b],
+                    outputs=[cond_img_b, segm_img_b],
+                    examples_per_page=4,
+                    fn=on_image_upload,
+                    run_on_click=True
+                )
+
+            with gr.Column():
+                # Description
+                gr.Markdown(description_b)
+
+    if args.enable_everything_mode:
+        with gr.Tab("Everything mode") as tab_e:
+            # Images
+            with gr.Row(variant="panel"):
+                with gr.Column(scale=1):
+                    cond_img_e.render()
+
+                with gr.Column(scale=1):
+                    segm_img_e.render()
+
+            # Submit & Clear
+            with gr.Row():
+                with gr.Column():
+                    with gr.Row():
+                        with gr.Column():
+                            segment_btn_e = gr.Button(
+                                "Start", variant="primary"
+                            )
+                            reset_btn_e = gr.Button("Reset", variant="secondary")
+
+                    gr.Markdown("Try some of the examples below ⬇️")
+                    gr.Examples(
+                        examples=examples,
+                        inputs=[cond_img_e],
+                        examples_per_page=4,
+                    )
+
+                with gr.Column():
+                    # Description
+                    gr.Markdown(description_e)
+
+    cond_img_p.upload(on_image_upload, cond_img_p, [cond_img_p, segm_img_p])
+    cond_img_p.select(get_points_with_draw, [cond_img_p, add_or_remove], cond_img_p)
+    segment_btn_p.click(
+        segment_with_points, inputs=[cond_img_p], outputs=[cond_img_p, segm_img_p]
+    )
+    clear_btn_p.click(clear, outputs=[cond_img_p, segm_img_p])
+    reset_btn_p.click(reset, outputs=[cond_img_p, segm_img_p])
+    tab_p.select(fn=reset_all, outputs=all_outputs)
+
+    cond_img_b.select(get_box_with_draw, [cond_img_b], cond_img_b)
+    segment_btn_b.click(
+        segment_with_box, inputs=[cond_img_b], outputs=[cond_img_b, segm_img_b]
+    )
+    clear_btn_b.click(clear, outputs=[cond_img_b, segm_img_b])
+    reset_btn_b.click(reset, outputs=[cond_img_b, segm_img_b])
+    tab_b.select(fn=reset_all, outputs=all_outputs)
+
+    if args.enable_everything_mode:
+        segment_btn_e.click(
+            segment_everything, inputs=[cond_img_e], outputs=segm_img_e
+        )
+        reset_btn_e.click(reset, outputs=[cond_img_e, segm_img_e])
+        tab_e.select(fn=reset_all, outputs=all_outputs)
+
+demo.queue()
+# demo.launch(server_name=args.server_name, server_port=args.port)
+demo.launch()

requirements.txt

@@ -0,0 +1,4 @@
+torch
+torchvision
+opencv-python
+timm

segment_anything/__init__.py

@@ -0,0 +1,15 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+from .build_sam import (
+    build_sam,
+    build_sam_vit_h,
+    build_sam_vit_l,
+    build_sam_vit_b,
+    sam_model_registry,
+)
+from .predictor import SamPredictor
+from .automatic_mask_generator import SamAutomaticMaskGenerator

segment_anything/automatic_mask_generator.py

@@ -0,0 +1,372 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import numpy as np
+import torch
+from torchvision.ops.boxes import batched_nms, box_area  # type: ignore
+
+from typing import Any, Dict, List, Optional, Tuple
+
+from .modeling import Sam
+from .predictor import SamPredictor
+from .utils.amg import (
+    MaskData,
+    area_from_rle,
+    batch_iterator,
+    batched_mask_to_box,
+    box_xyxy_to_xywh,
+    build_all_layer_point_grids,
+    calculate_stability_score,
+    coco_encode_rle,
+    generate_crop_boxes,
+    is_box_near_crop_edge,
+    mask_to_rle_pytorch,
+    remove_small_regions,
+    rle_to_mask,
+    uncrop_boxes_xyxy,
+    uncrop_masks,
+    uncrop_points,
+)
+
+
+class SamAutomaticMaskGenerator:
+    def __init__(
+        self,
+        model: Sam,
+        points_per_side: Optional[int] = 32,
+        points_per_batch: int = 64,
+        pred_iou_thresh: float = 0.88,
+        stability_score_thresh: float = 0.95,
+        stability_score_offset: float = 1.0,
+        box_nms_thresh: float = 0.7,
+        crop_n_layers: int = 0,
+        crop_nms_thresh: float = 0.7,
+        crop_overlap_ratio: float = 512 / 1500,
+        crop_n_points_downscale_factor: int = 1,
+        point_grids: Optional[List[np.ndarray]] = None,
+        min_mask_region_area: int = 0,
+        output_mode: str = "binary_mask",
+    ) -> None:
+        """
+        Using a SAM model, generates masks for the entire image.
+        Generates a grid of point prompts over the image, then filters
+        low quality and duplicate masks. The default settings are chosen
+        for SAM with a ViT-H backbone.
+
+        Arguments:
+          model (Sam): The SAM model to use for mask prediction.
+          points_per_side (int or None): The number of points to be sampled
+            along one side of the image. The total number of points is
+            points_per_side**2. If None, 'point_grids' must provide explicit
+            point sampling.
+          points_per_batch (int): Sets the number of points run simultaneously
+            by the model. Higher numbers may be faster but use more GPU memory.
+          pred_iou_thresh (float): A filtering threshold in [0,1], using the
+            model's predicted mask quality.
+          stability_score_thresh (float): A filtering threshold in [0,1], using
+            the stability of the mask under changes to the cutoff used to binarize
+            the model's mask predictions.
+          stability_score_offset (float): The amount to shift the cutoff when
+            calculated the stability score.
+          box_nms_thresh (float): The box IoU cutoff used by non-maximal
+            suppression to filter duplicate masks.
+          crop_n_layers (int): If >0, mask prediction will be run again on
+            crops of the image. Sets the number of layers to run, where each
+            layer has 2**i_layer number of image crops.
+          crop_nms_thresh (float): The box IoU cutoff used by non-maximal
+            suppression to filter duplicate masks between different crops.
+          crop_overlap_ratio (float): Sets the degree to which crops overlap.
+            In the first crop layer, crops will overlap by this fraction of
+            the image length. Later layers with more crops scale down this overlap.
+          crop_n_points_downscale_factor (int): The number of points-per-side
+            sampled in layer n is scaled down by crop_n_points_downscale_factor**n.
+          point_grids (list(np.ndarray) or None): A list over explicit grids
+            of points used for sampling, normalized to [0,1]. The nth grid in the
+            list is used in the nth crop layer. Exclusive with points_per_side.
+          min_mask_region_area (int): If >0, postprocessing will be applied
+            to remove disconnected regions and holes in masks with area smaller
+            than min_mask_region_area. Requires opencv.
+          output_mode (str): The form masks are returned in. Can be 'binary_mask',
+            'uncompressed_rle', or 'coco_rle'. 'coco_rle' requires pycocotools.
+            For large resolutions, 'binary_mask' may consume large amounts of
+            memory.
+        """
+
+        assert (points_per_side is None) != (
+            point_grids is None
+        ), "Exactly one of points_per_side or point_grid must be provided."
+        if points_per_side is not None:
+            self.point_grids = build_all_layer_point_grids(
+                points_per_side,
+                crop_n_layers,
+                crop_n_points_downscale_factor,
+            )
+        elif point_grids is not None:
+            self.point_grids = point_grids
+        else:
+            raise ValueError("Can't have both points_per_side and point_grid be None.")
+
+        assert output_mode in [
+            "binary_mask",
+            "uncompressed_rle",
+            "coco_rle",
+        ], f"Unknown output_mode {output_mode}."
+        if output_mode == "coco_rle":
+            from pycocotools import mask as mask_utils  # type: ignore # noqa: F401
+
+        if min_mask_region_area > 0:
+            import cv2  # type: ignore # noqa: F401
+
+        self.predictor = SamPredictor(model)
+        self.points_per_batch = points_per_batch
+        self.pred_iou_thresh = pred_iou_thresh
+        self.stability_score_thresh = stability_score_thresh
+        self.stability_score_offset = stability_score_offset
+        self.box_nms_thresh = box_nms_thresh
+        self.crop_n_layers = crop_n_layers
+        self.crop_nms_thresh = crop_nms_thresh
+        self.crop_overlap_ratio = crop_overlap_ratio
+        self.crop_n_points_downscale_factor = crop_n_points_downscale_factor
+        self.min_mask_region_area = min_mask_region_area
+        self.output_mode = output_mode
+
+    @torch.no_grad()
+    def generate(self, image: np.ndarray) -> List[Dict[str, Any]]:
+        """
+        Generates masks for the given image.
+
+        Arguments:
+          image (np.ndarray): The image to generate masks for, in HWC uint8 format.
+
+        Returns:
+           list(dict(str, any)): A list over records for masks. Each record is
+             a dict containing the following keys:
+               segmentation (dict(str, any) or np.ndarray): The mask. If
+                 output_mode='binary_mask', is an array of shape HW. Otherwise,
+                 is a dictionary containing the RLE.
+               bbox (list(float)): The box around the mask, in XYWH format.
+               area (int): The area in pixels of the mask.
+               predicted_iou (float): The model's own prediction of the mask's
+                 quality. This is filtered by the pred_iou_thresh parameter.
+               point_coords (list(list(float))): The point coordinates input
+                 to the model to generate this mask.
+               stability_score (float): A measure of the mask's quality. This
+                 is filtered on using the stability_score_thresh parameter.
+               crop_box (list(float)): The crop of the image used to generate
+                 the mask, given in XYWH format.
+        """
+
+        # Generate masks
+        mask_data = self._generate_masks(image)
+
+        # Filter small disconnected regions and holes in masks
+        if self.min_mask_region_area > 0:
+            mask_data = self.postprocess_small_regions(
+                mask_data,
+                self.min_mask_region_area,
+                max(self.box_nms_thresh, self.crop_nms_thresh),
+            )
+
+        # Encode masks
+        if self.output_mode == "coco_rle":
+            mask_data["segmentations"] = [coco_encode_rle(rle) for rle in mask_data["rles"]]
+        elif self.output_mode == "binary_mask":
+            mask_data["segmentations"] = [rle_to_mask(rle) for rle in mask_data["rles"]]
+        else:
+            mask_data["segmentations"] = mask_data["rles"]
+
+        # Write mask records
+        curr_anns = []
+        for idx in range(len(mask_data["segmentations"])):
+            ann = {
+                "segmentation": mask_data["segmentations"][idx],
+                "area": area_from_rle(mask_data["rles"][idx]),
+                "bbox": box_xyxy_to_xywh(mask_data["boxes"][idx]).tolist(),
+                "predicted_iou": mask_data["iou_preds"][idx].item(),
+                "point_coords": [mask_data["points"][idx].tolist()],
+                "stability_score": mask_data["stability_score"][idx].item(),
+                "crop_box": box_xyxy_to_xywh(mask_data["crop_boxes"][idx]).tolist(),
+            }
+            curr_anns.append(ann)
+
+        return curr_anns
+
+    def _generate_masks(self, image: np.ndarray) -> MaskData:
+        orig_size = image.shape[:2]
+        crop_boxes, layer_idxs = generate_crop_boxes(
+            orig_size, self.crop_n_layers, self.crop_overlap_ratio
+        )
+
+        # Iterate over image crops
+        data = MaskData()
+        for crop_box, layer_idx in zip(crop_boxes, layer_idxs):
+            crop_data = self._process_crop(image, crop_box, layer_idx, orig_size)
+            data.cat(crop_data)
+
+        # Remove duplicate masks between crops
+        if len(crop_boxes) > 1:
+            # Prefer masks from smaller crops
+            scores = 1 / box_area(data["crop_boxes"])
+            scores = scores.to(data["boxes"].device)
+            keep_by_nms = batched_nms(
+                data["boxes"].float(),
+                scores,
+                torch.zeros_like(data["boxes"][:, 0]),  # categories
+                iou_threshold=self.crop_nms_thresh,
+            )
+            data.filter(keep_by_nms)
+
+        data.to_numpy()
+        return data
+
+    def _process_crop(
+        self,
+        image: np.ndarray,
+        crop_box: List[int],
+        crop_layer_idx: int,
+        orig_size: Tuple[int, ...],
+    ) -> MaskData:
+        # Crop the image and calculate embeddings
+        x0, y0, x1, y1 = crop_box
+        cropped_im = image[y0:y1, x0:x1, :]
+        cropped_im_size = cropped_im.shape[:2]
+        self.predictor.set_image(cropped_im)
+
+        # Get points for this crop
+        points_scale = np.array(cropped_im_size)[None, ::-1]
+        points_for_image = self.point_grids[crop_layer_idx] * points_scale
+
+        # Generate masks for this crop in batches
+        data = MaskData()
+        for (points,) in batch_iterator(self.points_per_batch, points_for_image):
+            batch_data = self._process_batch(points, cropped_im_size, crop_box, orig_size)
+            data.cat(batch_data)
+            del batch_data
+        self.predictor.reset_image()
+
+        # Remove duplicates within this crop.
+        keep_by_nms = batched_nms(
+            data["boxes"].float(),
+            data["iou_preds"],
+            torch.zeros_like(data["boxes"][:, 0]),  # categories
+            iou_threshold=self.box_nms_thresh,
+        )
+        data.filter(keep_by_nms)
+
+        # Return to the original image frame
+        data["boxes"] = uncrop_boxes_xyxy(data["boxes"], crop_box)
+        data["points"] = uncrop_points(data["points"], crop_box)
+        data["crop_boxes"] = torch.tensor([crop_box for _ in range(len(data["rles"]))])
+
+        return data
+
+    def _process_batch(
+        self,
+        points: np.ndarray,
+        im_size: Tuple[int, ...],
+        crop_box: List[int],
+        orig_size: Tuple[int, ...],
+    ) -> MaskData:
+        orig_h, orig_w = orig_size
+
+        # Run model on this batch
+        transformed_points = self.predictor.transform.apply_coords(points, im_size)
+        in_points = torch.as_tensor(transformed_points, device=self.predictor.device)
+        in_labels = torch.ones(in_points.shape[0], dtype=torch.int, device=in_points.device)
+        masks, iou_preds, _ = self.predictor.predict_torch(
+            in_points[:, None, :],
+            in_labels[:, None],
+            num_multimask_outputs=3,
+            return_logits=True,
+        )
+
+        # Serialize predictions and store in MaskData
+        data = MaskData(
+            masks=masks.flatten(0, 1),
+            iou_preds=iou_preds.flatten(0, 1),
+            points=torch.as_tensor(points.repeat(masks.shape[1], axis=0)),
+        )
+        del masks
+
+        # Filter by predicted IoU
+        if self.pred_iou_thresh > 0.0:
+            keep_mask = data["iou_preds"] > self.pred_iou_thresh
+            data.filter(keep_mask)
+
+        # Calculate stability score
+        data["stability_score"] = calculate_stability_score(
+            data["masks"], self.predictor.model.mask_threshold, self.stability_score_offset
+        )
+        if self.stability_score_thresh > 0.0:
+            keep_mask = data["stability_score"] >= self.stability_score_thresh
+            data.filter(keep_mask)
+
+        # Threshold masks and calculate boxes
+        data["masks"] = data["masks"] > self.predictor.model.mask_threshold
+        data["boxes"] = batched_mask_to_box(data["masks"])
+
+        # Filter boxes that touch crop boundaries
+        keep_mask = ~is_box_near_crop_edge(data["boxes"], crop_box, [0, 0, orig_w, orig_h])
+        if not torch.all(keep_mask):
+            data.filter(keep_mask)
+
+        # Compress to RLE
+        data["masks"] = uncrop_masks(data["masks"], crop_box, orig_h, orig_w)
+        data["rles"] = mask_to_rle_pytorch(data["masks"])
+        del data["masks"]
+
+        return data
+
+    @staticmethod
+    def postprocess_small_regions(
+        mask_data: MaskData, min_area: int, nms_thresh: float
+    ) -> MaskData:
+        """
+        Removes small disconnected regions and holes in masks, then reruns
+        box NMS to remove any new duplicates.
+
+        Edits mask_data in place.
+
+        Requires open-cv as a dependency.
+        """
+        if len(mask_data["rles"]) == 0:
+            return mask_data
+
+        # Filter small disconnected regions and holes
+        new_masks = []
+        scores = []
+        for rle in mask_data["rles"]:
+            mask = rle_to_mask(rle)
+
+            mask, changed = remove_small_regions(mask, min_area, mode="holes")
+            unchanged = not changed
+            mask, changed = remove_small_regions(mask, min_area, mode="islands")
+            unchanged = unchanged and not changed
+
+            new_masks.append(torch.as_tensor(mask).unsqueeze(0))
+            # Give score=0 to changed masks and score=1 to unchanged masks
+            # so NMS will prefer ones that didn't need postprocessing
+            scores.append(float(unchanged))
+
+        # Recalculate boxes and remove any new duplicates
+        masks = torch.cat(new_masks, dim=0)
+        boxes = batched_mask_to_box(masks)
+        keep_by_nms = batched_nms(
+            boxes.float(),
+            torch.as_tensor(scores),
+            torch.zeros_like(boxes[:, 0]),  # categories
+            iou_threshold=nms_thresh,
+        )
+
+        # Only recalculate RLEs for masks that have changed
+        for i_mask in keep_by_nms:
+            if scores[i_mask] == 0.0:
+                mask_torch = masks[i_mask].unsqueeze(0)
+                mask_data["rles"][i_mask] = mask_to_rle_pytorch(mask_torch)[0]
+                mask_data["boxes"][i_mask] = boxes[i_mask]  # update res directly
+        mask_data.filter(keep_by_nms)
+
+        return mask_data

segment_anything/build_sam.py

@@ -0,0 +1,124 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import torch
+
+from functools import partial
+
+from .modeling import ImageEncoderViT, MaskDecoder, PromptEncoder, Sam, TwoWayTransformer, RepViT
+
+
+prompt_embed_dim = 256
+image_size = 1024
+vit_patch_size = 16
+image_embedding_size = image_size // vit_patch_size
+
+
+def build_sam_vit_h(checkpoint=None):
+    image_encoder = _build_sam_encoder(
+        encoder_embed_dim=1280,
+        encoder_depth=32,
+        encoder_num_heads=16,
+        encoder_global_attn_indexes=[7, 15, 23, 31]
+    )
+    return _build_sam(image_encoder, checkpoint)
+
+
+def build_sam_vit_l(checkpoint=None):
+    image_encoder = _build_sam_encoder(
+        encoder_embed_dim=1024,
+        encoder_depth=24,
+        encoder_num_heads=16,
+        encoder_global_attn_indexes=[5, 11, 17, 23]
+    )
+    return _build_sam(image_encoder, checkpoint)
+
+
+def build_sam_vit_b(checkpoint=None):
+    image_encoder = _build_sam_encoder(
+        encoder_embed_dim=768,
+        encoder_depth=12,
+        encoder_num_heads=12,
+        encoder_global_attn_indexes=[2, 5, 8, 11]
+    )
+    return _build_sam(image_encoder, checkpoint)
+
+
+def build_edge_sam(checkpoint=None, upsample_mode="bicubic"):
+    image_encoder = RepViT(
+        arch="m1",
+        img_size=image_size,
+        upsample_mode=upsample_mode
+    )
+    return _build_sam(image_encoder, checkpoint)
+
+
+sam_model_registry = {
+    "default": build_edge_sam,
+    "vit_h": build_sam_vit_h,
+    "vit_l": build_sam_vit_l,
+    "vit_b": build_sam_vit_b,
+    "edge_sam": build_edge_sam,
+}
+build_sam = build_edge_sam
+
+
+def _build_sam_encoder(
+    encoder_embed_dim,
+    encoder_depth,
+    encoder_num_heads,
+    encoder_global_attn_indexes,
+):
+    image_encoder = ImageEncoderViT(
+        depth=encoder_depth,
+        embed_dim=encoder_embed_dim,
+        img_size=image_size,
+        mlp_ratio=4,
+        norm_layer=partial(torch.nn.LayerNorm, eps=1e-6),
+        num_heads=encoder_num_heads,
+        patch_size=vit_patch_size,
+        qkv_bias=True,
+        use_rel_pos=True,
+        global_attn_indexes=encoder_global_attn_indexes,
+        window_size=14,
+        out_chans=prompt_embed_dim,
+    )
+    return image_encoder
+
+
+def _build_sam(
+    image_encoder,
+    checkpoint=None,
+):
+    sam = Sam(
+        image_encoder=image_encoder,
+        prompt_encoder=PromptEncoder(
+            embed_dim=prompt_embed_dim,
+            image_embedding_size=(image_embedding_size, image_embedding_size),
+            input_image_size=(image_size, image_size),
+            mask_in_chans=16,
+        ),
+        mask_decoder=MaskDecoder(
+            num_multimask_outputs=3,
+            transformer=TwoWayTransformer(
+                depth=2,
+                embedding_dim=prompt_embed_dim,
+                mlp_dim=2048,
+                num_heads=8,
+            ),
+            transformer_dim=prompt_embed_dim,
+            iou_head_depth=3,
+            iou_head_hidden_dim=256,
+        ),
+        pixel_mean=[123.675, 116.28, 103.53],
+        pixel_std=[58.395, 57.12, 57.375],
+    )
+    sam.eval()
+    if checkpoint is not None:
+        with open(checkpoint, "rb") as f:
+            state_dict = torch.load(f, map_location="cpu")
+        sam.load_state_dict(state_dict)
+    return sam

segment_anything/modeling/__init__.py

@@ -0,0 +1,12 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+from .sam import Sam
+from .image_encoder import ImageEncoderViT
+from .mask_decoder import MaskDecoder
+from .prompt_encoder import PromptEncoder
+from .transformer import TwoWayTransformer
+from .rep_vit import *

segment_anything/modeling/common.py

@@ -0,0 +1,118 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from typing import Type
+
+
+class MLPBlock(nn.Module):
+    def __init__(
+        self,
+        embedding_dim: int,
+        mlp_dim: int,
+        act: Type[nn.Module] = nn.GELU,
+    ) -> None:
+        super().__init__()
+        self.lin1 = nn.Linear(embedding_dim, mlp_dim)
+        self.lin2 = nn.Linear(mlp_dim, embedding_dim)
+        self.act = act()
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.lin2(self.act(self.lin1(x)))
+
+
+# From https://github.com/facebookresearch/detectron2/blob/main/detectron2/layers/batch_norm.py # noqa
+# Itself from https://github.com/facebookresearch/ConvNeXt/blob/d1fa8f6fef0a165b27399986cc2bdacc92777e40/models/convnext.py#L119  # noqa
+class LayerNorm2d(nn.Module):
+    def __init__(self, num_channels: int, eps: float = 1e-6) -> None:
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(num_channels))
+        self.bias = nn.Parameter(torch.zeros(num_channels))
+        self.eps = eps
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        u = x.mean(1, keepdim=True)
+        s = (x - u).pow(2).mean(1, keepdim=True)
+        x = (x - u) / torch.sqrt(s + self.eps)
+        x = self.weight[:, None, None] * x + self.bias[:, None, None]
+        return x
+
+
+def val2list(x: list or tuple or any, repeat_time=1) -> list:
+    if isinstance(x, (list, tuple)):
+        return list(x)
+    return [x for _ in range(repeat_time)]
+
+
+def val2tuple(x: list or tuple or any, min_len: int = 1, idx_repeat: int = -1) -> tuple:
+    x = val2list(x)
+
+    # repeat elements if necessary
+    if len(x) > 0:
+        x[idx_repeat:idx_repeat] = [x[idx_repeat] for _ in range(min_len - len(x))]
+
+    return tuple(x)
+
+
+def list_sum(x: list) -> any:
+    return x[0] if len(x) == 1 else x[0] + list_sum(x[1:])
+
+
+def resize(
+        x: torch.Tensor,
+        size: any or None = None,
+        scale_factor=None,
+        mode: str = "bicubic",
+        align_corners: bool or None = False,
+) -> torch.Tensor:
+    if mode in ["bilinear", "bicubic"]:
+        return F.interpolate(
+            x,
+            size=size,
+            scale_factor=scale_factor,
+            mode=mode,
+            align_corners=align_corners,
+        )
+    elif mode in ["nearest", "area"]:
+        return F.interpolate(x, size=size, scale_factor=scale_factor, mode=mode)
+    else:
+        raise NotImplementedError(f"resize(mode={mode}) not implemented.")
+
+
+class UpSampleLayer(nn.Module):
+    def __init__(
+            self,
+            mode="bicubic",
+            size=None,
+            factor=2,
+            align_corners=False,
+    ):
+        super(UpSampleLayer, self).__init__()
+        self.mode = mode
+        self.size = val2list(size, 2) if size is not None else None
+        self.factor = None if self.size is not None else factor
+        self.align_corners = align_corners
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return resize(x, self.size, self.factor, self.mode, self.align_corners)
+
+
+class OpSequential(nn.Module):
+    def __init__(self, op_list):
+        super(OpSequential, self).__init__()
+        valid_op_list = []
+        for op in op_list:
+            if op is not None:
+                valid_op_list.append(op)
+        self.op_list = nn.ModuleList(valid_op_list)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for op in self.op_list:
+            x = op(x)
+        return x

segment_anything/modeling/image_encoder.py

@@ -0,0 +1,395 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from typing import Optional, Tuple, Type
+
+from .common import LayerNorm2d, MLPBlock
+
+
+# This class and its supporting functions below lightly adapted from the ViTDet backbone available at: https://github.com/facebookresearch/detectron2/blob/main/detectron2/modeling/backbone/vit.py # noqa
+class ImageEncoderViT(nn.Module):
+    def __init__(
+        self,
+        img_size: int = 1024,
+        patch_size: int = 16,
+        in_chans: int = 3,
+        embed_dim: int = 768,
+        depth: int = 12,
+        num_heads: int = 12,
+        mlp_ratio: float = 4.0,
+        out_chans: int = 256,
+        qkv_bias: bool = True,
+        norm_layer: Type[nn.Module] = nn.LayerNorm,
+        act_layer: Type[nn.Module] = nn.GELU,
+        use_abs_pos: bool = True,
+        use_rel_pos: bool = False,
+        rel_pos_zero_init: bool = True,
+        window_size: int = 0,
+        global_attn_indexes: Tuple[int, ...] = (),
+    ) -> None:
+        """
+        Args:
+            img_size (int): Input image size.
+            patch_size (int): Patch size.
+            in_chans (int): Number of input image channels.
+            embed_dim (int): Patch embedding dimension.
+            depth (int): Depth of ViT.
+            num_heads (int): Number of attention heads in each ViT block.
+            mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
+            qkv_bias (bool): If True, add a learnable bias to query, key, value.
+            norm_layer (nn.Module): Normalization layer.
+            act_layer (nn.Module): Activation layer.
+            use_abs_pos (bool): If True, use absolute positional embeddings.
+            use_rel_pos (bool): If True, add relative positional embeddings to the attention map.
+            rel_pos_zero_init (bool): If True, zero initialize relative positional parameters.
+            window_size (int): Window size for window attention blocks.
+            global_attn_indexes (list): Indexes for blocks using global attention.
+        """
+        super().__init__()
+        self.img_size = img_size
+
+        self.patch_embed = PatchEmbed(
+            kernel_size=(patch_size, patch_size),
+            stride=(patch_size, patch_size),
+            in_chans=in_chans,
+            embed_dim=embed_dim,
+        )
+
+        self.pos_embed: Optional[nn.Parameter] = None
+        if use_abs_pos:
+            # Initialize absolute positional embedding with pretrain image size.
+            self.pos_embed = nn.Parameter(
+                torch.zeros(1, img_size // patch_size, img_size // patch_size, embed_dim)
+            )
+
+        self.blocks = nn.ModuleList()
+        for i in range(depth):
+            block = Block(
+                dim=embed_dim,
+                num_heads=num_heads,
+                mlp_ratio=mlp_ratio,
+                qkv_bias=qkv_bias,
+                norm_layer=norm_layer,
+                act_layer=act_layer,
+                use_rel_pos=use_rel_pos,
+                rel_pos_zero_init=rel_pos_zero_init,
+                window_size=window_size if i not in global_attn_indexes else 0,
+                input_size=(img_size // patch_size, img_size // patch_size),
+            )
+            self.blocks.append(block)
+
+        self.neck = nn.Sequential(
+            nn.Conv2d(
+                embed_dim,
+                out_chans,
+                kernel_size=1,
+                bias=False,
+            ),
+            LayerNorm2d(out_chans),
+            nn.Conv2d(
+                out_chans,
+                out_chans,
+                kernel_size=3,
+                padding=1,
+                bias=False,
+            ),
+            LayerNorm2d(out_chans),
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = self.patch_embed(x)
+        if self.pos_embed is not None:
+            x = x + self.pos_embed
+
+        for blk in self.blocks:
+            x = blk(x)
+
+        x = self.neck(x.permute(0, 3, 1, 2))
+
+        return x
+
+
+class Block(nn.Module):
+    """Transformer blocks with support of window attention and residual propagation blocks"""
+
+    def __init__(
+        self,
+        dim: int,
+        num_heads: int,
+        mlp_ratio: float = 4.0,
+        qkv_bias: bool = True,
+        norm_layer: Type[nn.Module] = nn.LayerNorm,
+        act_layer: Type[nn.Module] = nn.GELU,
+        use_rel_pos: bool = False,
+        rel_pos_zero_init: bool = True,
+        window_size: int = 0,
+        input_size: Optional[Tuple[int, int]] = None,
+    ) -> None:
+        """
+        Args:
+            dim (int): Number of input channels.
+            num_heads (int): Number of attention heads in each ViT block.
+            mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
+            qkv_bias (bool): If True, add a learnable bias to query, key, value.
+            norm_layer (nn.Module): Normalization layer.
+            act_layer (nn.Module): Activation layer.
+            use_rel_pos (bool): If True, add relative positional embeddings to the attention map.
+            rel_pos_zero_init (bool): If True, zero initialize relative positional parameters.
+            window_size (int): Window size for window attention blocks. If it equals 0, then
+                use global attention.
+            input_size (tuple(int, int) or None): Input resolution for calculating the relative
+                positional parameter size.
+        """
+        super().__init__()
+        self.norm1 = norm_layer(dim)
+        self.attn = Attention(
+            dim,
+            num_heads=num_heads,
+            qkv_bias=qkv_bias,
+            use_rel_pos=use_rel_pos,
+            rel_pos_zero_init=rel_pos_zero_init,
+            input_size=input_size if window_size == 0 else (window_size, window_size),
+        )
+
+        self.norm2 = norm_layer(dim)
+        self.mlp = MLPBlock(embedding_dim=dim, mlp_dim=int(dim * mlp_ratio), act=act_layer)
+
+        self.window_size = window_size
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        shortcut = x
+        x = self.norm1(x)
+        # Window partition
+        if self.window_size > 0:
+            H, W = x.shape[1], x.shape[2]
+            x, pad_hw = window_partition(x, self.window_size)
+
+        x = self.attn(x)
+        # Reverse window partition
+        if self.window_size > 0:
+            x = window_unpartition(x, self.window_size, pad_hw, (H, W))
+
+        x = shortcut + x
+        x = x + self.mlp(self.norm2(x))
+
+        return x
+
+
+class Attention(nn.Module):
+    """Multi-head Attention block with relative position embeddings."""
+
+    def __init__(
+        self,
+        dim: int,
+        num_heads: int = 8,
+        qkv_bias: bool = True,
+        use_rel_pos: bool = False,
+        rel_pos_zero_init: bool = True,
+        input_size: Optional[Tuple[int, int]] = None,
+    ) -> None:
+        """
+        Args:
+            dim (int): Number of input channels.
+            num_heads (int): Number of attention heads.
+            qkv_bias (bool):  If True, add a learnable bias to query, key, value.
+            rel_pos (bool): If True, add relative positional embeddings to the attention map.
+            rel_pos_zero_init (bool): If True, zero initialize relative positional parameters.
+            input_size (tuple(int, int) or None): Input resolution for calculating the relative
+                positional parameter size.
+        """
+        super().__init__()
+        self.num_heads = num_heads
+        head_dim = dim // num_heads
+        self.scale = head_dim**-0.5
+
+        self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
+        self.proj = nn.Linear(dim, dim)
+
+        self.use_rel_pos = use_rel_pos
+        if self.use_rel_pos:
+            assert (
+                input_size is not None
+            ), "Input size must be provided if using relative positional encoding."
+            # initialize relative positional embeddings
+            self.rel_pos_h = nn.Parameter(torch.zeros(2 * input_size[0] - 1, head_dim))
+            self.rel_pos_w = nn.Parameter(torch.zeros(2 * input_size[1] - 1, head_dim))
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        B, H, W, _ = x.shape
+        # qkv with shape (3, B, nHead, H * W, C)
+        qkv = self.qkv(x).reshape(B, H * W, 3, self.num_heads, -1).permute(2, 0, 3, 1, 4)
+        # q, k, v with shape (B * nHead, H * W, C)
+        q, k, v = qkv.reshape(3, B * self.num_heads, H * W, -1).unbind(0)
+
+        attn = (q * self.scale) @ k.transpose(-2, -1)
+
+        if self.use_rel_pos:
+            attn = add_decomposed_rel_pos(attn, q, self.rel_pos_h, self.rel_pos_w, (H, W), (H, W))
+
+        attn = attn.softmax(dim=-1)
+        x = (attn @ v).view(B, self.num_heads, H, W, -1).permute(0, 2, 3, 1, 4).reshape(B, H, W, -1)
+        x = self.proj(x)
+
+        return x
+
+
+def window_partition(x: torch.Tensor, window_size: int) -> Tuple[torch.Tensor, Tuple[int, int]]:
+    """
+    Partition into non-overlapping windows with padding if needed.
+    Args:
+        x (tensor): input tokens with [B, H, W, C].
+        window_size (int): window size.
+
+    Returns:
+        windows: windows after partition with [B * num_windows, window_size, window_size, C].
+        (Hp, Wp): padded height and width before partition
+    """
+    B, H, W, C = x.shape
+
+    pad_h = (window_size - H % window_size) % window_size
+    pad_w = (window_size - W % window_size) % window_size
+    if pad_h > 0 or pad_w > 0:
+        x = F.pad(x, (0, 0, 0, pad_w, 0, pad_h))
+    Hp, Wp = H + pad_h, W + pad_w
+
+    x = x.view(B, Hp // window_size, window_size, Wp // window_size, window_size, C)
+    windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size, window_size, C)
+    return windows, (Hp, Wp)
+
+
+def window_unpartition(
+    windows: torch.Tensor, window_size: int, pad_hw: Tuple[int, int], hw: Tuple[int, int]
+) -> torch.Tensor:
+    """
+    Window unpartition into original sequences and removing padding.
+    Args:
+        windows (tensor): input tokens with [B * num_windows, window_size, window_size, C].
+        window_size (int): window size.
+        pad_hw (Tuple): padded height and width (Hp, Wp).
+        hw (Tuple): original height and width (H, W) before padding.
+
+    Returns:
+        x: unpartitioned sequences with [B, H, W, C].
+    """
+    Hp, Wp = pad_hw
+    H, W = hw
+    B = windows.shape[0] // (Hp * Wp // window_size // window_size)
+    x = windows.view(B, Hp // window_size, Wp // window_size, window_size, window_size, -1)
+    x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(B, Hp, Wp, -1)
+
+    if Hp > H or Wp > W:
+        x = x[:, :H, :W, :].contiguous()
+    return x
+
+
+def get_rel_pos(q_size: int, k_size: int, rel_pos: torch.Tensor) -> torch.Tensor:
+    """
+    Get relative positional embeddings according to the relative positions of
+        query and key sizes.
+    Args:
+        q_size (int): size of query q.
+        k_size (int): size of key k.
+        rel_pos (Tensor): relative position embeddings (L, C).
+
+    Returns:
+        Extracted positional embeddings according to relative positions.
+    """
+    max_rel_dist = int(2 * max(q_size, k_size) - 1)
+    # Interpolate rel pos if needed.
+    if rel_pos.shape[0] != max_rel_dist:
+        # Interpolate rel pos.
+        rel_pos_resized = F.interpolate(
+            rel_pos.reshape(1, rel_pos.shape[0], -1).permute(0, 2, 1),
+            size=max_rel_dist,
+            mode="linear",
+        )
+        rel_pos_resized = rel_pos_resized.reshape(-1, max_rel_dist).permute(1, 0)
+    else:
+        rel_pos_resized = rel_pos
+
+    # Scale the coords with short length if shapes for q and k are different.
+    q_coords = torch.arange(q_size)[:, None] * max(k_size / q_size, 1.0)
+    k_coords = torch.arange(k_size)[None, :] * max(q_size / k_size, 1.0)
+    relative_coords = (q_coords - k_coords) + (k_size - 1) * max(q_size / k_size, 1.0)
+
+    return rel_pos_resized[relative_coords.long()]
+
+
+def add_decomposed_rel_pos(
+    attn: torch.Tensor,
+    q: torch.Tensor,
+    rel_pos_h: torch.Tensor,
+    rel_pos_w: torch.Tensor,
+    q_size: Tuple[int, int],
+    k_size: Tuple[int, int],
+) -> torch.Tensor:
+    """
+    Calculate decomposed Relative Positional Embeddings from :paper:`mvitv2`.
+    https://github.com/facebookresearch/mvit/blob/19786631e330df9f3622e5402b4a419a263a2c80/mvit/models/attention.py   # noqa B950
+    Args:
+        attn (Tensor): attention map.
+        q (Tensor): query q in the attention layer with shape (B, q_h * q_w, C).
+        rel_pos_h (Tensor): relative position embeddings (Lh, C) for height axis.
+        rel_pos_w (Tensor): relative position embeddings (Lw, C) for width axis.
+        q_size (Tuple): spatial sequence size of query q with (q_h, q_w).
+        k_size (Tuple): spatial sequence size of key k with (k_h, k_w).
+
+    Returns:
+        attn (Tensor): attention map with added relative positional embeddings.
+    """
+    q_h, q_w = q_size
+    k_h, k_w = k_size
+    Rh = get_rel_pos(q_h, k_h, rel_pos_h)
+    Rw = get_rel_pos(q_w, k_w, rel_pos_w)
+
+    B, _, dim = q.shape
+    r_q = q.reshape(B, q_h, q_w, dim)
+    rel_h = torch.einsum("bhwc,hkc->bhwk", r_q, Rh)
+    rel_w = torch.einsum("bhwc,wkc->bhwk", r_q, Rw)
+
+    attn = (
+        attn.view(B, q_h, q_w, k_h, k_w) + rel_h[:, :, :, :, None] + rel_w[:, :, :, None, :]
+    ).view(B, q_h * q_w, k_h * k_w)
+
+    return attn
+
+
+class PatchEmbed(nn.Module):
+    """
+    Image to Patch Embedding.
+    """
+
+    def __init__(
+        self,
+        kernel_size: Tuple[int, int] = (16, 16),
+        stride: Tuple[int, int] = (16, 16),
+        padding: Tuple[int, int] = (0, 0),
+        in_chans: int = 3,
+        embed_dim: int = 768,
+    ) -> None:
+        """
+        Args:
+            kernel_size (Tuple): kernel size of the projection layer.
+            stride (Tuple): stride of the projection layer.
+            padding (Tuple): padding size of the projection layer.
+            in_chans (int): Number of input image channels.
+            embed_dim (int): Patch embedding dimension.
+        """
+        super().__init__()
+
+        self.proj = nn.Conv2d(
+            in_chans, embed_dim, kernel_size=kernel_size, stride=stride, padding=padding
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = self.proj(x)
+        # B C H W -> B H W C
+        x = x.permute(0, 2, 3, 1)
+        return x

segment_anything/modeling/mask_decoder.py

@@ -0,0 +1,181 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+from typing import List, Tuple, Type
+
+from .common import LayerNorm2d
+
+
+class MaskDecoder(nn.Module):
+    def __init__(
+        self,
+        *,
+        transformer_dim: int,
+        transformer: nn.Module,
+        num_multimask_outputs: int = 3,
+        activation: Type[nn.Module] = nn.GELU,
+        iou_head_depth: int = 3,
+        iou_head_hidden_dim: int = 256,
+    ) -> None:
+        """
+        Predicts masks given an image and prompt embeddings, using a
+        transformer architecture.
+
+        Arguments:
+          transformer_dim (int): the channel dimension of the transformer
+          transformer (nn.Module): the transformer used to predict masks
+          num_multimask_outputs (int): the number of masks to predict
+            when disambiguating masks
+          activation (nn.Module): the type of activation to use when
+            upscaling masks
+          iou_head_depth (int): the depth of the MLP used to predict
+            mask quality
+          iou_head_hidden_dim (int): the hidden dimension of the MLP
+            used to predict mask quality
+        """
+        super().__init__()
+        self.transformer_dim = transformer_dim
+        self.transformer = transformer
+
+        self.num_multimask_outputs = num_multimask_outputs
+
+        self.iou_token = nn.Embedding(1, transformer_dim)
+        self.num_mask_tokens = num_multimask_outputs + 1
+        self.mask_tokens = nn.Embedding(self.num_mask_tokens, transformer_dim)
+
+        self.output_upscaling = nn.Sequential(
+            nn.ConvTranspose2d(transformer_dim, transformer_dim // 4, kernel_size=2, stride=2),
+            LayerNorm2d(transformer_dim // 4),
+            activation(),
+            nn.ConvTranspose2d(transformer_dim // 4, transformer_dim // 8, kernel_size=2, stride=2),
+            activation(),
+        )
+        self.output_hypernetworks_mlps = nn.ModuleList(
+            [
+                MLP(transformer_dim, transformer_dim, transformer_dim // 8, 3)
+                for i in range(self.num_mask_tokens)
+            ]
+        )
+
+        self.iou_prediction_head = MLP(
+            transformer_dim, iou_head_hidden_dim, self.num_mask_tokens, iou_head_depth
+        )
+
+    def forward(
+        self,
+        image_embeddings: torch.Tensor,
+        image_pe: torch.Tensor,
+        sparse_prompt_embeddings: torch.Tensor,
+        dense_prompt_embeddings: torch.Tensor,
+        num_multimask_outputs: int,
+    ) -> Tuple[torch.Tensor, torch.Tensor]:
+        """
+        Predict masks given image and prompt embeddings.
+
+        Arguments:
+          image_embeddings (torch.Tensor): the embeddings from the image encoder
+          image_pe (torch.Tensor): positional encoding with the shape of image_embeddings
+          sparse_prompt_embeddings (torch.Tensor): the embeddings of the points and boxes
+          dense_prompt_embeddings (torch.Tensor): the embeddings of the mask inputs
+          num_multimask_outputs (int): the number of masks to predict
+            when disambiguating masks
+
+        Returns:
+          torch.Tensor: batched predicted masks
+          torch.Tensor: batched predictions of mask quality
+        """
+        masks, iou_pred = self.predict_masks(
+            image_embeddings=image_embeddings,
+            image_pe=image_pe,
+            sparse_prompt_embeddings=sparse_prompt_embeddings,
+            dense_prompt_embeddings=dense_prompt_embeddings,
+        )
+
+        # Select the correct mask or masks for output
+        if num_multimask_outputs == 4:
+            mask_slice = slice(0, None)
+        elif num_multimask_outputs == 3:
+            mask_slice = slice(1, None)
+        elif num_multimask_outputs == 1:
+            mask_slice = slice(0, 1)
+        else:
+            raise ValueError
+
+        masks = masks[:, mask_slice, :, :]
+        iou_pred = iou_pred[:, mask_slice]
+
+        # Prepare output
+        return masks, iou_pred
+
+    def predict_masks(
+        self,
+        image_embeddings: torch.Tensor,
+        image_pe: torch.Tensor,
+        sparse_prompt_embeddings: torch.Tensor,
+        dense_prompt_embeddings: torch.Tensor,
+    ) -> Tuple[torch.Tensor, torch.Tensor]:
+        """Predicts masks. See 'forward' for more details."""
+        # Concatenate output tokens
+        output_tokens = torch.cat([self.iou_token.weight, self.mask_tokens.weight], dim=0)
+        output_tokens = output_tokens.unsqueeze(0).expand(sparse_prompt_embeddings.size(0), -1, -1)
+        tokens = torch.cat((output_tokens, sparse_prompt_embeddings), dim=1)
+
+        # Expand per-image data in batch direction to be per-mask
+        src = torch.repeat_interleave(image_embeddings, tokens.shape[0], dim=0)
+        src = src + dense_prompt_embeddings
+        pos_src = torch.repeat_interleave(image_pe, tokens.shape[0], dim=0)
+        b, c, h, w = src.shape
+
+        # Run the transformer
+        hs, src = self.transformer(src, pos_src, tokens)
+        iou_token_out = hs[:, 0, :]
+        mask_tokens_out = hs[:, 1 : (1 + self.num_mask_tokens), :]
+
+        # Upscale mask embeddings and predict masks using the mask tokens
+        src = src.transpose(1, 2).view(b, c, h, w)
+        upscaled_embedding = self.output_upscaling(src)
+        hyper_in_list: List[torch.Tensor] = []
+        for i in range(self.num_mask_tokens):
+            hyper_in_list.append(self.output_hypernetworks_mlps[i](mask_tokens_out[:, i, :]))
+        hyper_in = torch.stack(hyper_in_list, dim=1)
+        b, c, h, w = upscaled_embedding.shape
+        masks = (hyper_in @ upscaled_embedding.view(b, c, h * w)).view(b, -1, h, w)
+
+        # Generate mask quality predictions
+        iou_pred = self.iou_prediction_head(iou_token_out)
+
+        return masks, iou_pred
+
+
+# Lightly adapted from
+# https://github.com/facebookresearch/MaskFormer/blob/main/mask_former/modeling/transformer/transformer_predictor.py # noqa
+class MLP(nn.Module):
+    def __init__(
+        self,
+        input_dim: int,
+        hidden_dim: int,
+        output_dim: int,
+        num_layers: int,
+        sigmoid_output: bool = False,
+    ) -> None:
+        super().__init__()
+        self.num_layers = num_layers
+        h = [hidden_dim] * (num_layers - 1)
+        self.layers = nn.ModuleList(
+            nn.Linear(n, k) for n, k in zip([input_dim] + h, h + [output_dim])
+        )
+        self.sigmoid_output = sigmoid_output
+
+    def forward(self, x):
+        for i, layer in enumerate(self.layers):
+            x = F.relu(layer(x)) if i < self.num_layers - 1 else layer(x)
+        if self.sigmoid_output:
+            x = F.sigmoid(x)
+        return x

segment_anything/modeling/prompt_encoder.py

@@ -0,0 +1,214 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import numpy as np
+import torch
+from torch import nn
+
+from typing import Any, Optional, Tuple, Type
+
+from .common import LayerNorm2d
+
+
+class PromptEncoder(nn.Module):
+    def __init__(
+        self,
+        embed_dim: int,
+        image_embedding_size: Tuple[int, int],
+        input_image_size: Tuple[int, int],
+        mask_in_chans: int,
+        activation: Type[nn.Module] = nn.GELU,
+    ) -> None:
+        """
+        Encodes prompts for input to SAM's mask decoder.
+
+        Arguments:
+          embed_dim (int): The prompts' embedding dimension
+          image_embedding_size (tuple(int, int)): The spatial size of the
+            image embedding, as (H, W).
+          input_image_size (int): The padded size of the image as input
+            to the image encoder, as (H, W).
+          mask_in_chans (int): The number of hidden channels used for
+            encoding input masks.
+          activation (nn.Module): The activation to use when encoding
+            input masks.
+        """
+        super().__init__()
+        self.embed_dim = embed_dim
+        self.input_image_size = input_image_size
+        self.image_embedding_size = image_embedding_size
+        self.pe_layer = PositionEmbeddingRandom(embed_dim // 2)
+
+        self.num_point_embeddings: int = 4  # pos/neg point + 2 box corners
+        point_embeddings = [nn.Embedding(1, embed_dim) for i in range(self.num_point_embeddings)]
+        self.point_embeddings = nn.ModuleList(point_embeddings)
+        self.not_a_point_embed = nn.Embedding(1, embed_dim)
+
+        self.mask_input_size = (4 * image_embedding_size[0], 4 * image_embedding_size[1])
+        self.mask_downscaling = nn.Sequential(
+            nn.Conv2d(1, mask_in_chans // 4, kernel_size=2, stride=2),
+            LayerNorm2d(mask_in_chans // 4),
+            activation(),
+            nn.Conv2d(mask_in_chans // 4, mask_in_chans, kernel_size=2, stride=2),
+            LayerNorm2d(mask_in_chans),
+            activation(),
+            nn.Conv2d(mask_in_chans, embed_dim, kernel_size=1),
+        )
+        self.no_mask_embed = nn.Embedding(1, embed_dim)
+
+    def get_dense_pe(self) -> torch.Tensor:
+        """
+        Returns the positional encoding used to encode point prompts,
+        applied to a dense set of points the shape of the image encoding.
+
+        Returns:
+          torch.Tensor: Positional encoding with shape
+            1x(embed_dim)x(embedding_h)x(embedding_w)
+        """
+        return self.pe_layer(self.image_embedding_size).unsqueeze(0)
+
+    def _embed_points(
+        self,
+        points: torch.Tensor,
+        labels: torch.Tensor,
+        pad: bool,
+    ) -> torch.Tensor:
+        """Embeds point prompts."""
+        points = points + 0.5  # Shift to center of pixel
+        if pad:
+            padding_point = torch.zeros((points.shape[0], 1, 2), device=points.device)
+            padding_label = -torch.ones((labels.shape[0], 1), device=labels.device)
+            points = torch.cat([points, padding_point], dim=1)
+            labels = torch.cat([labels, padding_label], dim=1)
+        point_embedding = self.pe_layer.forward_with_coords(points, self.input_image_size)
+        point_embedding[labels == -1] = 0.0
+        point_embedding[labels == -1] += self.not_a_point_embed.weight
+        point_embedding[labels == 0] += self.point_embeddings[0].weight
+        point_embedding[labels == 1] += self.point_embeddings[1].weight
+        return point_embedding
+
+    def _embed_boxes(self, boxes: torch.Tensor) -> torch.Tensor:
+        """Embeds box prompts."""
+        boxes = boxes + 0.5  # Shift to center of pixel
+        coords = boxes.reshape(-1, 2, 2)
+        corner_embedding = self.pe_layer.forward_with_coords(coords, self.input_image_size)
+        corner_embedding[:, 0, :] += self.point_embeddings[2].weight
+        corner_embedding[:, 1, :] += self.point_embeddings[3].weight
+        return corner_embedding
+
+    def _embed_masks(self, masks: torch.Tensor) -> torch.Tensor:
+        """Embeds mask inputs."""
+        mask_embedding = self.mask_downscaling(masks)
+        return mask_embedding
+
+    def _get_batch_size(
+        self,
+        points: Optional[Tuple[torch.Tensor, torch.Tensor]],
+        boxes: Optional[torch.Tensor],
+        masks: Optional[torch.Tensor],
+    ) -> int:
+        """
+        Gets the batch size of the output given the batch size of the input prompts.
+        """
+        if points is not None:
+            return points[0].shape[0]
+        elif boxes is not None:
+            return boxes.shape[0]
+        elif masks is not None:
+            return masks.shape[0]
+        else:
+            return 1
+
+    def _get_device(self) -> torch.device:
+        return self.point_embeddings[0].weight.device
+
+    def forward(
+        self,
+        points: Optional[Tuple[torch.Tensor, torch.Tensor]],
+        boxes: Optional[torch.Tensor],
+        masks: Optional[torch.Tensor],
+    ) -> Tuple[torch.Tensor, torch.Tensor]:
+        """
+        Embeds different types of prompts, returning both sparse and dense
+        embeddings.
+
+        Arguments:
+          points (tuple(torch.Tensor, torch.Tensor) or none): point coordinates
+            and labels to embed.
+          boxes (torch.Tensor or none): boxes to embed
+          masks (torch.Tensor or none): masks to embed
+
+        Returns:
+          torch.Tensor: sparse embeddings for the points and boxes, with shape
+            BxNx(embed_dim), where N is determined by the number of input points
+            and boxes.
+          torch.Tensor: dense embeddings for the masks, in the shape
+            Bx(embed_dim)x(embed_H)x(embed_W)
+        """
+        bs = self._get_batch_size(points, boxes, masks)
+        sparse_embeddings = torch.empty((bs, 0, self.embed_dim), device=self._get_device())
+        if points is not None:
+            coords, labels = points
+            point_embeddings = self._embed_points(coords, labels, pad=(boxes is None))
+            sparse_embeddings = torch.cat([sparse_embeddings, point_embeddings], dim=1)
+        if boxes is not None:
+            box_embeddings = self._embed_boxes(boxes)
+            sparse_embeddings = torch.cat([sparse_embeddings, box_embeddings], dim=1)
+
+        if masks is not None:
+            dense_embeddings = self._embed_masks(masks)
+        else:
+            dense_embeddings = self.no_mask_embed.weight.reshape(1, -1, 1, 1).expand(
+                bs, -1, self.image_embedding_size[0], self.image_embedding_size[1]
+            )
+
+        return sparse_embeddings, dense_embeddings
+
+
+class PositionEmbeddingRandom(nn.Module):
+    """
+    Positional encoding using random spatial frequencies.
+    """
+
+    def __init__(self, num_pos_feats: int = 64, scale: Optional[float] = None) -> None:
+        super().__init__()
+        if scale is None or scale <= 0.0:
+            scale = 1.0
+        self.register_buffer(
+            "positional_encoding_gaussian_matrix",
+            scale * torch.randn((2, num_pos_feats)),
+        )
+
+    def _pe_encoding(self, coords: torch.Tensor) -> torch.Tensor:
+        """Positionally encode points that are normalized to [0,1]."""
+        # assuming coords are in [0, 1]^2 square and have d_1 x ... x d_n x 2 shape
+        coords = 2 * coords - 1
+        coords = coords @ self.positional_encoding_gaussian_matrix
+        coords = 2 * np.pi * coords
+        # outputs d_1 x ... x d_n x C shape
+        return torch.cat([torch.sin(coords), torch.cos(coords)], dim=-1)
+
+    def forward(self, size: Tuple[int, int]) -> torch.Tensor:
+        """Generate positional encoding for a grid of the specified size."""
+        h, w = size
+        device: Any = self.positional_encoding_gaussian_matrix.device
+        grid = torch.ones((h, w), device=device, dtype=torch.float32)
+        y_embed = grid.cumsum(dim=0) - 0.5
+        x_embed = grid.cumsum(dim=1) - 0.5
+        y_embed = y_embed / h
+        x_embed = x_embed / w
+
+        pe = self._pe_encoding(torch.stack([x_embed, y_embed], dim=-1))
+        return pe.permute(2, 0, 1)  # C x H x W
+
+    def forward_with_coords(
+        self, coords_input: torch.Tensor, image_size: Tuple[int, int]
+    ) -> torch.Tensor:
+        """Positionally encode points that are not normalized to [0,1]."""
+        coords = coords_input.clone()
+        coords[:, :, 0] = coords[:, :, 0] / image_size[1]
+        coords[:, :, 1] = coords[:, :, 1] / image_size[0]
+        return self._pe_encoding(coords.to(torch.float))  # B x N x C

segment_anything/modeling/rep_vit.py

@@ -0,0 +1,367 @@
+import torch.nn as nn
+from segment_anything.modeling.common import LayerNorm2d, UpSampleLayer, OpSequential
+
+__all__ = ['rep_vit_m1', 'rep_vit_m2', 'rep_vit_m3', 'RepViT']
+
+m1_cfgs = [
+    # k, t, c, SE, HS, s
+    [3, 2, 48, 1, 0, 1],
+    [3, 2, 48, 0, 0, 1],
+    [3, 2, 48, 0, 0, 1],
+    [3, 2, 96, 0, 0, 2],
+    [3, 2, 96, 1, 0, 1],
+    [3, 2, 96, 0, 0, 1],
+    [3, 2, 96, 0, 0, 1],
+    [3, 2, 192, 0, 1, 2],
+    [3, 2, 192, 1, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 192, 1, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 192, 1, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 192, 1, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 192, 1, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 192, 1, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 192, 1, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 192, 0, 1, 1],
+    [3, 2, 384, 0, 1, 2],
+    [3, 2, 384, 1, 1, 1],
+    [3, 2, 384, 0, 1, 1]
+]
+
+m2_cfgs = [
+    # k, t, c, SE, HS, s
+    [3, 2, 64, 1, 0, 1],
+    [3, 2, 64, 0, 0, 1],
+    [3, 2, 64, 0, 0, 1],
+    [3, 2, 128, 0, 0, 2],
+    [3, 2, 128, 1, 0, 1],
+    [3, 2, 128, 0, 0, 1],
+    [3, 2, 128, 0, 0, 1],
+    [3, 2, 256, 0, 1, 2],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 512, 0, 1, 2],
+    [3, 2, 512, 1, 1, 1],
+    [3, 2, 512, 0, 1, 1]
+]
+
+m3_cfgs = [
+    # k, t, c, SE, HS, s
+    [3, 2, 64, 1, 0, 1],
+    [3, 2, 64, 0, 0, 1],
+    [3, 2, 64, 1, 0, 1],
+    [3, 2, 64, 0, 0, 1],
+    [3, 2, 64, 0, 0, 1],
+    [3, 2, 128, 0, 0, 2],
+    [3, 2, 128, 1, 0, 1],
+    [3, 2, 128, 0, 0, 1],
+    [3, 2, 128, 1, 0, 1],
+    [3, 2, 128, 0, 0, 1],
+    [3, 2, 128, 0, 0, 1],
+    [3, 2, 256, 0, 1, 2],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 1, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 256, 0, 1, 1],
+    [3, 2, 512, 0, 1, 2],
+    [3, 2, 512, 1, 1, 1],
+    [3, 2, 512, 0, 1, 1]
+]
+
+
+def _make_divisible(v, divisor, min_value=None):
+    """
+    This function is taken from the original tf repo.
+    It ensures that all layers have a channel number that is divisible by 8
+    It can be seen here:
+    https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet/mobilenet.py
+    :param v:
+    :param divisor:
+    :param min_value:
+    :return:
+    """
+    if min_value is None:
+        min_value = divisor
+    new_v = max(min_value, int(v + divisor / 2) // divisor * divisor)
+    # Make sure that round down does not go down by more than 10%.
+    if new_v < 0.9 * v:
+        new_v += divisor
+    return new_v
+
+
+from timm.models.layers import SqueezeExcite
+
+import torch
+
+
+class Conv2d_BN(torch.nn.Sequential):
+    def __init__(self, a, b, ks=1, stride=1, pad=0, dilation=1,
+                 groups=1, bn_weight_init=1, resolution=-10000):
+        super().__init__()
+        self.add_module('c', torch.nn.Conv2d(
+            a, b, ks, stride, pad, dilation, groups, bias=False))
+        self.add_module('bn', torch.nn.BatchNorm2d(b))
+        torch.nn.init.constant_(self.bn.weight, bn_weight_init)
+        torch.nn.init.constant_(self.bn.bias, 0)
+
+    @torch.no_grad()
+    def fuse(self):
+        c, bn = self._modules.values()
+        w = bn.weight / (bn.running_var + bn.eps) ** 0.5
+        w = c.weight * w[:, None, None, None]
+        b = bn.bias - bn.running_mean * bn.weight / \
+            (bn.running_var + bn.eps) ** 0.5
+        m = torch.nn.Conv2d(w.size(1) * self.c.groups, w.size(
+            0), w.shape[2:], stride=self.c.stride, padding=self.c.padding, dilation=self.c.dilation,
+                            groups=self.c.groups,
+                            device=c.weight.device)
+        m.weight.data.copy_(w)
+        m.bias.data.copy_(b)
+        return m
+
+
+class Residual(torch.nn.Module):
+    def __init__(self, m, drop=0.):
+        super().__init__()
+        self.m = m
+        self.drop = drop
+
+    def forward(self, x):
+        if self.training and self.drop > 0:
+            return x + self.m(x) * torch.rand(x.size(0), 1, 1, 1,
+                                              device=x.device).ge_(self.drop).div(1 - self.drop).detach()
+        else:
+            return x + self.m(x)
+
+    @torch.no_grad()
+    def fuse(self):
+        if isinstance(self.m, Conv2d_BN):
+            m = self.m.fuse()
+            assert (m.groups == m.in_channels)
+            identity = torch.ones(m.weight.shape[0], m.weight.shape[1], 1, 1)
+            identity = torch.nn.functional.pad(identity, [1, 1, 1, 1])
+            m.weight += identity.to(m.weight.device)
+            return m
+        elif isinstance(self.m, torch.nn.Conv2d):
+            m = self.m
+            assert (m.groups != m.in_channels)
+            identity = torch.ones(m.weight.shape[0], m.weight.shape[1], 1, 1)
+            identity = torch.nn.functional.pad(identity, [1, 1, 1, 1])
+            m.weight += identity.to(m.weight.device)
+            return m
+        else:
+            return self
+
+
+class RepVGGDW(torch.nn.Module):
+    def __init__(self, ed) -> None:
+        super().__init__()
+        self.conv = Conv2d_BN(ed, ed, 3, 1, 1, groups=ed)
+        self.conv1 = Conv2d_BN(ed, ed, 1, 1, 0, groups=ed)
+        self.dim = ed
+
+    def forward(self, x):
+        return self.conv(x) + self.conv1(x) + x
+
+    @torch.no_grad()
+    def fuse(self):
+        conv = self.conv.fuse()
+        conv1 = self.conv1.fuse()
+
+        conv_w = conv.weight
+        conv_b = conv.bias
+        conv1_w = conv1.weight
+        conv1_b = conv1.bias
+
+        conv1_w = torch.nn.functional.pad(conv1_w, [1, 1, 1, 1])
+
+        identity = torch.nn.functional.pad(torch.ones(conv1_w.shape[0], conv1_w.shape[1], 1, 1, device=conv1_w.device),
+                                           [1, 1, 1, 1])
+
+        final_conv_w = conv_w + conv1_w + identity
+        final_conv_b = conv_b + conv1_b
+
+        conv.weight.data.copy_(final_conv_w)
+        conv.bias.data.copy_(final_conv_b)
+        return conv
+
+
+class RepViTBlock(nn.Module):
+    def __init__(self, inp, hidden_dim, oup, kernel_size, stride, use_se, use_hs, skip_downsample=False):
+        super(RepViTBlock, self).__init__()
+        assert stride in [1, 2]
+
+        self.identity = stride == 1 and inp == oup
+        assert (hidden_dim == 2 * inp)
+
+        if stride == 2:
+            if skip_downsample:
+                stride = 1
+            self.token_mixer = nn.Sequential(
+                Conv2d_BN(inp, inp, kernel_size, stride, (kernel_size - 1) // 2, groups=inp),
+                SqueezeExcite(inp, 0.25) if use_se else nn.Identity(),
+                Conv2d_BN(inp, oup, ks=1, stride=1, pad=0)
+            )
+            self.channel_mixer = Residual(nn.Sequential(
+                # pw
+                Conv2d_BN(oup, 2 * oup, 1, 1, 0),
+                nn.GELU() if use_hs else nn.GELU(),
+                # pw-linear
+                Conv2d_BN(2 * oup, oup, 1, 1, 0, bn_weight_init=0),
+            ))
+        else:
+            assert (self.identity)
+            self.token_mixer = nn.Sequential(
+                RepVGGDW(inp),
+                SqueezeExcite(inp, 0.25) if use_se else nn.Identity(),
+            )
+            self.channel_mixer = Residual(nn.Sequential(
+                # pw
+                Conv2d_BN(inp, hidden_dim, 1, 1, 0),
+                nn.GELU() if use_hs else nn.GELU(),
+                # pw-linear
+                Conv2d_BN(hidden_dim, oup, 1, 1, 0, bn_weight_init=0),
+            ))
+
+    def forward(self, x):
+        return self.channel_mixer(self.token_mixer(x))
+
+
+from timm.models.vision_transformer import trunc_normal_
+
+
+class BN_Linear(torch.nn.Sequential):
+    def __init__(self, a, b, bias=True, std=0.02):
+        super().__init__()
+        self.add_module('bn', torch.nn.BatchNorm1d(a))
+        self.add_module('l', torch.nn.Linear(a, b, bias=bias))
+        trunc_normal_(self.l.weight, std=std)
+        if bias:
+            torch.nn.init.constant_(self.l.bias, 0)
+
+    @torch.no_grad()
+    def fuse(self):
+        bn, l = self._modules.values()
+        w = bn.weight / (bn.running_var + bn.eps) ** 0.5
+        b = bn.bias - self.bn.running_mean * \
+            self.bn.weight / (bn.running_var + bn.eps) ** 0.5
+        w = l.weight * w[None, :]
+        if l.bias is None:
+            b = b @ self.l.weight.T
+        else:
+            b = (l.weight @ b[:, None]).view(-1) + self.l.bias
+        m = torch.nn.Linear(w.size(1), w.size(0), device=l.weight.device)
+        m.weight.data.copy_(w)
+        m.bias.data.copy_(b)
+        return m
+
+
+class RepViT(nn.Module):
+    arch_settings = {
+        'm1': m1_cfgs,
+        'm2': m2_cfgs,
+        'm3': m3_cfgs
+    }
+
+    def __init__(self, arch, img_size=1024, upsample_mode='bicubic'):
+        super(RepViT, self).__init__()
+        # setting of inverted residual blocks
+        self.cfgs = self.arch_settings[arch]
+        self.img_size = img_size
+
+        # building first layer
+        input_channel = self.cfgs[0][2]
+        patch_embed = torch.nn.Sequential(Conv2d_BN(3, input_channel // 2, 3, 2, 1), torch.nn.GELU(),
+                                          Conv2d_BN(input_channel // 2, input_channel, 3, 2, 1))
+        layers = [patch_embed]
+        # building inverted residual blocks
+        block = RepViTBlock
+        self.stage_idx = []
+        prev_c = input_channel
+        for idx, (k, t, c, use_se, use_hs, s) in enumerate(self.cfgs):
+            output_channel = _make_divisible(c, 8)
+            exp_size = _make_divisible(input_channel * t, 8)
+            skip_downsample = False
+            if c != prev_c:
+                self.stage_idx.append(idx - 1)
+                prev_c = c
+            layers.append(block(input_channel, exp_size, output_channel, k, s, use_se, use_hs, skip_downsample))
+            input_channel = output_channel
+        self.stage_idx.append(idx)
+        self.features = nn.ModuleList(layers)
+
+        stage2_channels = _make_divisible(self.cfgs[self.stage_idx[2]][2], 8)
+        stage3_channels = _make_divisible(self.cfgs[self.stage_idx[3]][2], 8)
+        self.fuse_stage2 = nn.Conv2d(stage2_channels, 256, kernel_size=1, bias=False)
+        self.fuse_stage3 = OpSequential([
+            nn.Conv2d(stage3_channels, 256, kernel_size=1, bias=False),
+            UpSampleLayer(factor=2, mode=upsample_mode),
+        ])
+
+        self.neck = nn.Sequential(
+            nn.Conv2d(256, 256, kernel_size=1, bias=False),
+            LayerNorm2d(256),
+            nn.Conv2d(256, 256, kernel_size=3, padding=1, bias=False),
+            LayerNorm2d(256),
+        )
+
+    def forward(self, x):
+        counter = 0
+        output_dict = dict()
+        # patch_embed
+        x = self.features[0](x)
+        output_dict['stem'] = x
+        # stages
+        for idx, f in enumerate(self.features[1:]):
+            x = f(x)
+            if idx in self.stage_idx:
+                output_dict[f'stage{counter}'] = x
+                counter += 1
+
+        x = self.fuse_stage2(output_dict['stage2']) + self.fuse_stage3(output_dict['stage3'])
+
+        x = self.neck(x)
+        return x
+
+
+def rep_vit_m1(img_size=1024, **kwargs):
+    return RepViT('m1', img_size, **kwargs)
+
+
+def rep_vit_m2(img_size=1024, **kwargs):
+    return RepViT('m2', img_size, **kwargs)
+
+
+def rep_vit_m3(img_size=1024, **kwargs):
+    return RepViT('m3', img_size, **kwargs)

segment_anything/modeling/sam.py

@@ -0,0 +1,187 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+from typing import Any, Dict, List, Tuple
+
+from .image_encoder import ImageEncoderViT
+from .mask_decoder import MaskDecoder
+from .prompt_encoder import PromptEncoder
+from ..utils.amg import calculate_stability_score
+
+
+class Sam(nn.Module):
+    mask_threshold: float = 0.0
+    image_format: str = "RGB"
+    stability_score_offset: float = 1.0
+
+    def __init__(
+        self,
+        image_encoder: ImageEncoderViT,
+        prompt_encoder: PromptEncoder,
+        mask_decoder: MaskDecoder,
+        pixel_mean: List[float] = [123.675, 116.28, 103.53],
+        pixel_std: List[float] = [58.395, 57.12, 57.375],
+    ) -> None:
+        """
+        SAM predicts object masks from an image and input prompts.
+
+        Arguments:
+          image_encoder (ImageEncoderViT): The backbone used to encode the
+            image into image embeddings that allow for efficient mask prediction.
+          prompt_encoder (PromptEncoder): Encodes various types of input prompts.
+          mask_decoder (MaskDecoder): Predicts masks from the image embeddings
+            and encoded prompts.
+          pixel_mean (list(float)): Mean values for normalizing pixels in the input image.
+          pixel_std (list(float)): Std values for normalizing pixels in the input image.
+        """
+        super().__init__()
+        self.image_encoder = image_encoder
+        self.prompt_encoder = prompt_encoder
+        self.mask_decoder = mask_decoder
+        self.register_buffer("pixel_mean", torch.Tensor(pixel_mean).view(-1, 1, 1), False)
+        self.register_buffer("pixel_std", torch.Tensor(pixel_std).view(-1, 1, 1), False)
+
+    @property
+    def device(self) -> Any:
+        return self.pixel_mean.device
+
+    @torch.no_grad()
+    def forward_dummy_encoder(self, x):
+        return self.image_encoder(x)
+
+    @torch.no_grad()
+    def forward(
+        self,
+        batched_input: List[Dict[str, Any]],
+        num_multimask_outputs: int = 1,
+        use_stability_score: bool = False
+    ) -> List[Dict[str, torch.Tensor]]:
+        """
+        Predicts masks end-to-end from provided images and prompts.
+        If prompts are not known in advance, using SamPredictor is
+        recommended over calling the model directly.
+
+        Arguments:
+          batched_input (list(dict)): A list over input images, each a
+            dictionary with the following keys. A prompt key can be
+            excluded if it is not present.
+              'image': The image as a torch tensor in 3xHxW format,
+                already transformed for input to the model.
+              'original_size': (tuple(int, int)) The original size of
+                the image before transformation, as (H, W).
+              'point_coords': (torch.Tensor) Batched point prompts for
+                this image, with shape BxNx2. Already transformed to the
+                input frame of the model.
+              'point_labels': (torch.Tensor) Batched labels for point prompts,
+                with shape BxN.
+              'boxes': (torch.Tensor) Batched box inputs, with shape Bx4.
+                Already transformed to the input frame of the model.
+              'mask_inputs': (torch.Tensor) Batched mask inputs to the model,
+                in the form Bx1xHxW.
+          num_multimask_outputs (int): the number of masks to predict
+            when disambiguating masks. Choices: 1, 3, 4.
+          use_stability_score (bool): If true, use stability scores to substitute
+            IoU predictions.
+
+        Returns:
+          (list(dict)): A list over input images, where each element is
+            as dictionary with the following keys.
+              'masks': (torch.Tensor) Batched binary mask predictions,
+                with shape BxCxHxW, where B is the number of input prompts,
+                C is determined by multimask_output, and (H, W) is the
+                original size of the image.
+              'iou_predictions': (torch.Tensor) The model's predictions
+                of mask quality, in shape BxC.
+              'low_res_logits': (torch.Tensor) Low resolution logits with
+                shape BxCxHxW, where H=W=256. Can be passed as mask input
+                to subsequent iterations of prediction.
+        """
+        input_images = torch.stack([self.preprocess(x["image"]) for x in batched_input], dim=0)
+        image_embeddings = self.image_encoder(input_images)
+
+        outputs = []
+        for image_record, curr_embedding in zip(batched_input, image_embeddings):
+            if "point_coords" in image_record:
+                points = (image_record["point_coords"], image_record["point_labels"])
+            else:
+                points = None
+            sparse_embeddings, dense_embeddings = self.prompt_encoder(
+                points=points,
+                boxes=image_record.get("boxes", None),
+                masks=image_record.get("mask_inputs", None),
+            )
+            low_res_masks, iou_predictions = self.mask_decoder(
+                image_embeddings=curr_embedding.unsqueeze(0),
+                image_pe=self.prompt_encoder.get_dense_pe(),
+                sparse_prompt_embeddings=sparse_embeddings,
+                dense_prompt_embeddings=dense_embeddings,
+                num_multimask_outputs=num_multimask_outputs,
+            )
+            if use_stability_score:
+                iou_predictions = calculate_stability_score(
+                    low_res_masks, self.mask_threshold, self.stability_score_offset
+                )
+            masks = self.postprocess_masks(
+                low_res_masks,
+                input_size=image_record["image"].shape[-2:],
+                original_size=image_record["original_size"],
+            )
+            masks = masks > self.mask_threshold
+            outputs.append(
+                {
+                    "masks": masks,
+                    "iou_predictions": iou_predictions,
+                    "low_res_logits": low_res_masks,
+                }
+            )
+        return outputs
+
+    def postprocess_masks(
+        self,
+        masks: torch.Tensor,
+        input_size: Tuple[int, ...],
+        original_size: Tuple[int, ...],
+    ) -> torch.Tensor:
+        """
+        Remove padding and upscale masks to the original image size.
+
+        Arguments:
+          masks (torch.Tensor): Batched masks from the mask_decoder,
+            in BxCxHxW format.
+          input_size (tuple(int, int)): The size of the image input to the
+            model, in (H, W) format. Used to remove padding.
+          original_size (tuple(int, int)): The original size of the image
+            before resizing for input to the model, in (H, W) format.
+
+        Returns:
+          (torch.Tensor): Batched masks in BxCxHxW format, where (H, W)
+            is given by original_size.
+        """
+        masks = F.interpolate(
+            masks,
+            (self.image_encoder.img_size, self.image_encoder.img_size),
+            mode="bilinear",
+            align_corners=False,
+        )
+        masks = masks[..., : input_size[0], : input_size[1]]
+        masks = F.interpolate(masks, original_size, mode="bilinear", align_corners=False)
+        return masks
+
+    def preprocess(self, x: torch.Tensor) -> torch.Tensor:
+        """Normalize pixel values and pad to a square input."""
+        # Normalize colors
+        x = (x - self.pixel_mean) / self.pixel_std
+
+        # Pad
+        h, w = x.shape[-2:]
+        padh = self.image_encoder.img_size - h
+        padw = self.image_encoder.img_size - w
+        x = F.pad(x, (0, padw, 0, padh))
+        return x

segment_anything/modeling/transformer.py

@@ -0,0 +1,240 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import torch
+from torch import Tensor, nn
+
+import math
+from typing import Tuple, Type
+
+from .common import MLPBlock
+
+
+class TwoWayTransformer(nn.Module):
+    def __init__(
+        self,
+        depth: int,
+        embedding_dim: int,
+        num_heads: int,
+        mlp_dim: int,
+        activation: Type[nn.Module] = nn.ReLU,
+        attention_downsample_rate: int = 2,
+    ) -> None:
+        """
+        A transformer decoder that attends to an input image using
+        queries whose positional embedding is supplied.
+
+        Args:
+          depth (int): number of layers in the transformer
+          embedding_dim (int): the channel dimension for the input embeddings
+          num_heads (int): the number of heads for multihead attention. Must
+            divide embedding_dim
+          mlp_dim (int): the channel dimension internal to the MLP block
+          activation (nn.Module): the activation to use in the MLP block
+        """
+        super().__init__()
+        self.depth = depth
+        self.embedding_dim = embedding_dim
+        self.num_heads = num_heads
+        self.mlp_dim = mlp_dim
+        self.layers = nn.ModuleList()
+
+        for i in range(depth):
+            self.layers.append(
+                TwoWayAttentionBlock(
+                    embedding_dim=embedding_dim,
+                    num_heads=num_heads,
+                    mlp_dim=mlp_dim,
+                    activation=activation,
+                    attention_downsample_rate=attention_downsample_rate,
+                    skip_first_layer_pe=(i == 0),
+                )
+            )
+
+        self.final_attn_token_to_image = Attention(
+            embedding_dim, num_heads, downsample_rate=attention_downsample_rate
+        )
+        self.norm_final_attn = nn.LayerNorm(embedding_dim)
+
+    def forward(
+        self,
+        image_embedding: Tensor,
+        image_pe: Tensor,
+        point_embedding: Tensor,
+    ) -> Tuple[Tensor, Tensor]:
+        """
+        Args:
+          image_embedding (torch.Tensor): image to attend to. Should be shape
+            B x embedding_dim x h x w for any h and w.
+          image_pe (torch.Tensor): the positional encoding to add to the image. Must
+            have the same shape as image_embedding.
+          point_embedding (torch.Tensor): the embedding to add to the query points.
+            Must have shape B x N_points x embedding_dim for any N_points.
+
+        Returns:
+          torch.Tensor: the processed point_embedding
+          torch.Tensor: the processed image_embedding
+        """
+        # BxCxHxW -> BxHWxC == B x N_image_tokens x C
+        bs, c, h, w = image_embedding.shape
+        image_embedding = image_embedding.flatten(2).permute(0, 2, 1)
+        image_pe = image_pe.flatten(2).permute(0, 2, 1)
+
+        # Prepare queries
+        queries = point_embedding
+        keys = image_embedding
+
+        # Apply transformer blocks and final layernorm
+        for layer in self.layers:
+            queries, keys = layer(
+                queries=queries,
+                keys=keys,
+                query_pe=point_embedding,
+                key_pe=image_pe,
+            )
+
+        # Apply the final attention layer from the points to the image
+        q = queries + point_embedding
+        k = keys + image_pe
+        attn_out = self.final_attn_token_to_image(q=q, k=k, v=keys)
+        queries = queries + attn_out
+        queries = self.norm_final_attn(queries)
+
+        return queries, keys
+
+
+class TwoWayAttentionBlock(nn.Module):
+    def __init__(
+        self,
+        embedding_dim: int,
+        num_heads: int,
+        mlp_dim: int = 2048,
+        activation: Type[nn.Module] = nn.ReLU,
+        attention_downsample_rate: int = 2,
+        skip_first_layer_pe: bool = False,
+    ) -> None:
+        """
+        A transformer block with four layers: (1) self-attention of sparse
+        inputs, (2) cross attention of sparse inputs to dense inputs, (3) mlp
+        block on sparse inputs, and (4) cross attention of dense inputs to sparse
+        inputs.
+
+        Arguments:
+          embedding_dim (int): the channel dimension of the embeddings
+          num_heads (int): the number of heads in the attention layers
+          mlp_dim (int): the hidden dimension of the mlp block
+          activation (nn.Module): the activation of the mlp block
+          skip_first_layer_pe (bool): skip the PE on the first layer
+        """
+        super().__init__()
+        self.self_attn = Attention(embedding_dim, num_heads)
+        self.norm1 = nn.LayerNorm(embedding_dim)
+
+        self.cross_attn_token_to_image = Attention(
+            embedding_dim, num_heads, downsample_rate=attention_downsample_rate
+        )
+        self.norm2 = nn.LayerNorm(embedding_dim)
+
+        self.mlp = MLPBlock(embedding_dim, mlp_dim, activation)
+        self.norm3 = nn.LayerNorm(embedding_dim)
+
+        self.norm4 = nn.LayerNorm(embedding_dim)
+        self.cross_attn_image_to_token = Attention(
+            embedding_dim, num_heads, downsample_rate=attention_downsample_rate
+        )
+
+        self.skip_first_layer_pe = skip_first_layer_pe
+
+    def forward(
+        self, queries: Tensor, keys: Tensor, query_pe: Tensor, key_pe: Tensor
+    ) -> Tuple[Tensor, Tensor]:
+        # Self attention block
+        if self.skip_first_layer_pe:
+            queries = self.self_attn(q=queries, k=queries, v=queries)
+        else:
+            q = queries + query_pe
+            attn_out = self.self_attn(q=q, k=q, v=queries)
+            queries = queries + attn_out
+        queries = self.norm1(queries)
+
+        # Cross attention block, tokens attending to image embedding
+        q = queries + query_pe
+        k = keys + key_pe
+        attn_out = self.cross_attn_token_to_image(q=q, k=k, v=keys)
+        queries = queries + attn_out
+        queries = self.norm2(queries)
+
+        # MLP block
+        mlp_out = self.mlp(queries)
+        queries = queries + mlp_out
+        queries = self.norm3(queries)
+
+        # Cross attention block, image embedding attending to tokens
+        q = queries + query_pe
+        k = keys + key_pe
+        attn_out = self.cross_attn_image_to_token(q=k, k=q, v=queries)
+        keys = keys + attn_out
+        keys = self.norm4(keys)
+
+        return queries, keys
+
+
+class Attention(nn.Module):
+    """
+    An attention layer that allows for downscaling the size of the embedding
+    after projection to queries, keys, and values.
+    """
+
+    def __init__(
+        self,
+        embedding_dim: int,
+        num_heads: int,
+        downsample_rate: int = 1,
+    ) -> None:
+        super().__init__()
+        self.embedding_dim = embedding_dim
+        self.internal_dim = embedding_dim // downsample_rate
+        self.num_heads = num_heads
+        assert self.internal_dim % num_heads == 0, "num_heads must divide embedding_dim."
+
+        self.q_proj = nn.Linear(embedding_dim, self.internal_dim)
+        self.k_proj = nn.Linear(embedding_dim, self.internal_dim)
+        self.v_proj = nn.Linear(embedding_dim, self.internal_dim)
+        self.out_proj = nn.Linear(self.internal_dim, embedding_dim)
+
+    def _separate_heads(self, x: Tensor, num_heads: int) -> Tensor:
+        b, n, c = x.shape
+        x = x.reshape(b, n, num_heads, c // num_heads)
+        return x.transpose(1, 2)  # B x N_heads x N_tokens x C_per_head
+
+    def _recombine_heads(self, x: Tensor) -> Tensor:
+        b, n_heads, n_tokens, c_per_head = x.shape
+        x = x.transpose(1, 2)
+        return x.reshape(b, n_tokens, n_heads * c_per_head)  # B x N_tokens x C
+
+    def forward(self, q: Tensor, k: Tensor, v: Tensor) -> Tensor:
+        # Input projections
+        q = self.q_proj(q)
+        k = self.k_proj(k)
+        v = self.v_proj(v)
+
+        # Separate into heads
+        q = self._separate_heads(q, self.num_heads)
+        k = self._separate_heads(k, self.num_heads)
+        v = self._separate_heads(v, self.num_heads)
+
+        # Attention
+        _, _, _, c_per_head = q.shape
+        attn = q @ k.permute(0, 1, 3, 2)  # B x N_heads x N_tokens x N_tokens
+        attn = attn / math.sqrt(c_per_head)
+        attn = torch.softmax(attn, dim=-1)
+
+        # Get output
+        out = attn @ v
+        out = self._recombine_heads(out)
+        out = self.out_proj(out)
+
+        return out

segment_anything/predictor.py

@@ -0,0 +1,275 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import numpy as np
+import torch
+
+from segment_anything.modeling import Sam
+
+from typing import Optional, Tuple
+
+from .utils.transforms import ResizeLongestSide
+from .utils.amg import calculate_stability_score
+
+
+class SamPredictor:
+    def __init__(
+        self,
+        sam_model: Sam,
+    ) -> None:
+        """
+        Uses SAM to calculate the image embedding for an image, and then
+        allow repeated, efficient mask prediction given prompts.
+
+        Arguments:
+          sam_model (Sam): The model to use for mask prediction.
+        """
+        super().__init__()
+        self.model = sam_model
+        self.transform = ResizeLongestSide(sam_model.image_encoder.img_size)
+        self.stability_score_offset = 1.0
+        self.reset_image()
+
+    def set_image(
+        self,
+        image: np.ndarray,
+        image_format: str = "RGB",
+    ) -> None:
+        """
+        Calculates the image embeddings for the provided image, allowing
+        masks to be predicted with the 'predict' method.
+
+        Arguments:
+          image (np.ndarray): The image for calculating masks. Expects an
+            image in HWC uint8 format, with pixel values in [0, 255].
+          image_format (str): The color format of the image, in ['RGB', 'BGR'].
+        """
+        assert image_format in [
+            "RGB",
+            "BGR",
+        ], f"image_format must be in ['RGB', 'BGR'], is {image_format}."
+        if image_format != self.model.image_format:
+            image = image[..., ::-1]
+
+        # Transform the image to the form expected by the model
+        input_image = self.transform.apply_image(image)
+        input_image_torch = torch.as_tensor(input_image, device=self.device)
+        input_image_torch = input_image_torch.permute(2, 0, 1).contiguous()[None, :, :, :]
+
+        self.set_torch_image(input_image_torch, image.shape[:2])
+
+    @torch.no_grad()
+    def set_torch_image(
+        self,
+        transformed_image: torch.Tensor,
+        original_image_size: Tuple[int, ...],
+    ) -> None:
+        """
+        Calculates the image embeddings for the provided image, allowing
+        masks to be predicted with the 'predict' method. Expects the input
+        image to be already transformed to the format expected by the model.
+
+        Arguments:
+          transformed_image (torch.Tensor): The input image, with shape
+            1x3xHxW, which has been transformed with ResizeLongestSide.
+          original_image_size (tuple(int, int)): The size of the image
+            before transformation, in (H, W) format.
+        """
+        assert (
+            len(transformed_image.shape) == 4
+            and transformed_image.shape[1] == 3
+            and max(*transformed_image.shape[2:]) == self.model.image_encoder.img_size
+        ), f"set_torch_image input must be BCHW with long side {self.model.image_encoder.img_size}."
+        self.reset_image()
+
+        self.original_size = original_image_size
+        self.input_size = tuple(transformed_image.shape[-2:])
+        input_image = self.model.preprocess(transformed_image)
+        self.features = self.model.image_encoder(input_image)
+        self.is_image_set = True
+
+    def predict(
+        self,
+        point_coords: Optional[np.ndarray] = None,
+        point_labels: Optional[np.ndarray] = None,
+        box: Optional[np.ndarray] = None,
+        mask_input: Optional[np.ndarray] = None,
+        num_multimask_outputs: int = 3,
+        return_logits: bool = False,
+        use_stability_score: bool = False
+    ) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
+        """
+        Predict masks for the given input prompts, using the currently set image.
+
+        Arguments:
+          point_coords (np.ndarray or None): A Nx2 array of point prompts to the
+            model. Each point is in (X,Y) in pixels.
+          point_labels (np.ndarray or None): A length N array of labels for the
+            point prompts. 1 indicates a foreground point and 0 indicates a
+            background point.
+          box (np.ndarray or None): A length 4 array given a box prompt to the
+            model, in XYXY format.
+          mask_input (np.ndarray): A low resolution mask input to the model, typically
+            coming from a previous prediction iteration. Has form 1xHxW, where
+            for SAM, H=W=256.
+          num_multimask_outputs (int): the number of masks to predict
+            when disambiguating masks. Choices: 1, 3, 4.
+          return_logits (bool): If true, returns un-thresholded masks logits
+            instead of a binary mask.
+          use_stability_score (bool): If true, use stability scores to substitute
+            IoU predictions.
+
+        Returns:
+          (np.ndarray): The output masks in CxHxW format, where C is the
+            number of masks, and (H, W) is the original image size.
+          (np.ndarray): An array of length C containing the model's
+            predictions for the quality of each mask.
+          (np.ndarray): An array of shape CxHxW, where C is the number
+            of masks and H=W=256. These low resolution logits can be passed to
+            a subsequent iteration as mask input.
+        """
+        if not self.is_image_set:
+            raise RuntimeError("An image must be set with .set_image(...) before mask prediction.")
+
+        # Transform input prompts
+        coords_torch, labels_torch, box_torch, mask_input_torch = None, None, None, None
+        if point_coords is not None:
+            assert (
+                point_labels is not None
+            ), "point_labels must be supplied if point_coords is supplied."
+            point_coords = self.transform.apply_coords(point_coords, self.original_size)
+            coords_torch = torch.as_tensor(point_coords, dtype=torch.float, device=self.device)
+            labels_torch = torch.as_tensor(point_labels, dtype=torch.int, device=self.device)
+            coords_torch, labels_torch = coords_torch[None, :, :], labels_torch[None, :]
+        if box is not None:
+            box = self.transform.apply_boxes(box, self.original_size)
+            box_torch = torch.as_tensor(box, dtype=torch.float, device=self.device)
+            box_torch = box_torch[None, :]
+        if mask_input is not None:
+            mask_input_torch = torch.as_tensor(mask_input, dtype=torch.float, device=self.device)
+            mask_input_torch = mask_input_torch[None, :, :, :]
+
+        masks, iou_predictions, low_res_masks = self.predict_torch(
+            coords_torch,
+            labels_torch,
+            box_torch,
+            mask_input_torch,
+            num_multimask_outputs,
+            return_logits=return_logits,
+            use_stability_score=use_stability_score
+        )
+
+        masks_np = masks[0].detach().cpu().numpy()
+        iou_predictions_np = iou_predictions[0].detach().cpu().numpy()
+        low_res_masks_np = low_res_masks[0].detach().cpu().numpy()
+        return masks_np, iou_predictions_np, low_res_masks_np
+
+    @torch.no_grad()
+    def predict_torch(
+        self,
+        point_coords: Optional[torch.Tensor],
+        point_labels: Optional[torch.Tensor],
+        boxes: Optional[torch.Tensor] = None,
+        mask_input: Optional[torch.Tensor] = None,
+        num_multimask_outputs: int = 3,
+        return_logits: bool = False,
+        use_stability_score: bool = True
+    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
+        """
+        Predict masks for the given input prompts, using the currently set image.
+        Input prompts are batched torch tensors and are expected to already be
+        transformed to the input frame using ResizeLongestSide.
+
+        Arguments:
+          point_coords (torch.Tensor or None): A BxNx2 array of point prompts to the
+            model. Each point is in (X,Y) in pixels.
+          point_labels (torch.Tensor or None): A BxN array of labels for the
+            point prompts. 1 indicates a foreground point and 0 indicates a
+            background point.
+          boxes (np.ndarray or None): A Bx4 array given a box prompt to the
+            model, in XYXY format.
+          mask_input (np.ndarray): A low resolution mask input to the model, typically
+            coming from a previous prediction iteration. Has form Bx1xHxW, where
+            for SAM, H=W=256. Masks returned by a previous iteration of the
+            predict method do not need further transformation.
+          num_multimask_outputs (int): the number of masks to predict
+            when disambiguating masks. Choices: 1, 3, 4.
+          return_logits (bool): If true, returns un-thresholded masks logits
+            instead of a binary mask.
+          use_stability_score (bool): If true, use stability scores to substitute
+            IoU predictions.
+
+        Returns:
+          (torch.Tensor): The output masks in BxCxHxW format, where C is the
+            number of masks, and (H, W) is the original image size.
+          (torch.Tensor): An array of shape BxC containing the model's
+            predictions for the quality of each mask.
+          (torch.Tensor): An array of shape BxCxHxW, where C is the number
+            of masks and H=W=256. These low res logits can be passed to
+            a subsequent iteration as mask input.
+        """
+        if not self.is_image_set:
+            raise RuntimeError("An image must be set with .set_image(...) before mask prediction.")
+
+        if point_coords is not None:
+            points = (point_coords, point_labels)
+        else:
+            points = None
+
+        # Embed prompts
+        sparse_embeddings, dense_embeddings = self.model.prompt_encoder(
+            points=points,
+            boxes=boxes,
+            masks=mask_input,
+        )
+
+        # Predict masks
+        low_res_masks, iou_predictions = self.model.mask_decoder(
+            image_embeddings=self.features,
+            image_pe=self.model.prompt_encoder.get_dense_pe(),
+            sparse_prompt_embeddings=sparse_embeddings,
+            dense_prompt_embeddings=dense_embeddings,
+            num_multimask_outputs=num_multimask_outputs,
+        )
+
+        if use_stability_score:
+            iou_predictions = calculate_stability_score(
+                low_res_masks, self.model.mask_threshold, self.stability_score_offset
+            )
+
+        # Upscale the masks to the original image resolution
+        masks = self.model.postprocess_masks(low_res_masks, self.input_size, self.original_size)
+
+        if not return_logits:
+            masks = masks > self.model.mask_threshold
+
+        return masks, iou_predictions, low_res_masks
+
+    def get_image_embedding(self) -> torch.Tensor:
+        """
+        Returns the image embeddings for the currently set image, with
+        shape 1xCxHxW, where C is the embedding dimension and (H,W) are
+        the embedding spatial dimension of SAM (typically C=256, H=W=64).
+        """
+        if not self.is_image_set:
+            raise RuntimeError(
+                "An image must be set with .set_image(...) to generate an embedding."
+            )
+        assert self.features is not None, "Features must exist if an image has been set."
+        return self.features
+
+    @property
+    def device(self) -> torch.device:
+        return self.model.device
+
+    def reset_image(self) -> None:
+        """Resets the currently set image."""
+        self.is_image_set = False
+        self.features = None
+        self.orig_h = None
+        self.orig_w = None
+        self.input_h = None
+        self.input_w = None

segment_anything/utils/__init__.py

@@ -0,0 +1,5 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.

segment_anything/utils/amg.py

@@ -0,0 +1,346 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import numpy as np
+import torch
+
+import math
+from copy import deepcopy
+from itertools import product
+from typing import Any, Dict, Generator, ItemsView, List, Tuple
+
+
+class MaskData:
+    """
+    A structure for storing masks and their related data in batched format.
+    Implements basic filtering and concatenation.
+    """
+
+    def __init__(self, **kwargs) -> None:
+        for v in kwargs.values():
+            assert isinstance(
+                v, (list, np.ndarray, torch.Tensor)
+            ), "MaskData only supports list, numpy arrays, and torch tensors."
+        self._stats = dict(**kwargs)
+
+    def __setitem__(self, key: str, item: Any) -> None:
+        assert isinstance(
+            item, (list, np.ndarray, torch.Tensor)
+        ), "MaskData only supports list, numpy arrays, and torch tensors."
+        self._stats[key] = item
+
+    def __delitem__(self, key: str) -> None:
+        del self._stats[key]
+
+    def __getitem__(self, key: str) -> Any:
+        return self._stats[key]
+
+    def items(self) -> ItemsView[str, Any]:
+        return self._stats.items()
+
+    def filter(self, keep: torch.Tensor) -> None:
+        for k, v in self._stats.items():
+            if v is None:
+                self._stats[k] = None
+            elif isinstance(v, torch.Tensor):
+                self._stats[k] = v[torch.as_tensor(keep, device=v.device)]
+            elif isinstance(v, np.ndarray):
+                self._stats[k] = v[keep.detach().cpu().numpy()]
+            elif isinstance(v, list) and keep.dtype == torch.bool:
+                self._stats[k] = [a for i, a in enumerate(v) if keep[i]]
+            elif isinstance(v, list):
+                self._stats[k] = [v[i] for i in keep]
+            else:
+                raise TypeError(f"MaskData key {k} has an unsupported type {type(v)}.")
+
+    def cat(self, new_stats: "MaskData") -> None:
+        for k, v in new_stats.items():
+            if k not in self._stats or self._stats[k] is None:
+                self._stats[k] = deepcopy(v)
+            elif isinstance(v, torch.Tensor):
+                self._stats[k] = torch.cat([self._stats[k], v], dim=0)
+            elif isinstance(v, np.ndarray):
+                self._stats[k] = np.concatenate([self._stats[k], v], axis=0)
+            elif isinstance(v, list):
+                self._stats[k] = self._stats[k] + deepcopy(v)
+            else:
+                raise TypeError(f"MaskData key {k} has an unsupported type {type(v)}.")
+
+    def to_numpy(self) -> None:
+        for k, v in self._stats.items():
+            if isinstance(v, torch.Tensor):
+                self._stats[k] = v.detach().cpu().numpy()
+
+
+def is_box_near_crop_edge(
+    boxes: torch.Tensor, crop_box: List[int], orig_box: List[int], atol: float = 20.0
+) -> torch.Tensor:
+    """Filter masks at the edge of a crop, but not at the edge of the original image."""
+    crop_box_torch = torch.as_tensor(crop_box, dtype=torch.float, device=boxes.device)
+    orig_box_torch = torch.as_tensor(orig_box, dtype=torch.float, device=boxes.device)
+    boxes = uncrop_boxes_xyxy(boxes, crop_box).float()
+    near_crop_edge = torch.isclose(boxes, crop_box_torch[None, :], atol=atol, rtol=0)
+    near_image_edge = torch.isclose(boxes, orig_box_torch[None, :], atol=atol, rtol=0)
+    near_crop_edge = torch.logical_and(near_crop_edge, ~near_image_edge)
+    return torch.any(near_crop_edge, dim=1)
+
+
+def box_xyxy_to_xywh(box_xyxy: torch.Tensor) -> torch.Tensor:
+    box_xywh = deepcopy(box_xyxy)
+    box_xywh[2] = box_xywh[2] - box_xywh[0]
+    box_xywh[3] = box_xywh[3] - box_xywh[1]
+    return box_xywh
+
+
+def batch_iterator(batch_size: int, *args) -> Generator[List[Any], None, None]:
+    assert len(args) > 0 and all(
+        len(a) == len(args[0]) for a in args
+    ), "Batched iteration must have inputs of all the same size."
+    n_batches = len(args[0]) // batch_size + int(len(args[0]) % batch_size != 0)
+    for b in range(n_batches):
+        yield [arg[b * batch_size : (b + 1) * batch_size] for arg in args]
+
+
+def mask_to_rle_pytorch(tensor: torch.Tensor) -> List[Dict[str, Any]]:
+    """
+    Encodes masks to an uncompressed RLE, in the format expected by
+    pycoco tools.
+    """
+    # Put in fortran order and flatten h,w
+    b, h, w = tensor.shape
+    tensor = tensor.permute(0, 2, 1).flatten(1)
+
+    # Compute change indices
+    diff = tensor[:, 1:] ^ tensor[:, :-1]
+    change_indices = diff.nonzero()
+
+    # Encode run length
+    out = []
+    for i in range(b):
+        cur_idxs = change_indices[change_indices[:, 0] == i, 1]
+        cur_idxs = torch.cat(
+            [
+                torch.tensor([0], dtype=cur_idxs.dtype, device=cur_idxs.device),
+                cur_idxs + 1,
+                torch.tensor([h * w], dtype=cur_idxs.dtype, device=cur_idxs.device),
+            ]
+        )
+        btw_idxs = cur_idxs[1:] - cur_idxs[:-1]
+        counts = [] if tensor[i, 0] == 0 else [0]
+        counts.extend(btw_idxs.detach().cpu().tolist())
+        out.append({"size": [h, w], "counts": counts})
+    return out
+
+
+def rle_to_mask(rle: Dict[str, Any]) -> np.ndarray:
+    """Compute a binary mask from an uncompressed RLE."""
+    h, w = rle["size"]
+    mask = np.empty(h * w, dtype=bool)
+    idx = 0
+    parity = False
+    for count in rle["counts"]:
+        mask[idx : idx + count] = parity
+        idx += count
+        parity ^= True
+    mask = mask.reshape(w, h)
+    return mask.transpose()  # Put in C order
+
+
+def area_from_rle(rle: Dict[str, Any]) -> int:
+    return sum(rle["counts"][1::2])
+
+
+def calculate_stability_score(
+    masks: torch.Tensor, mask_threshold: float, threshold_offset: float
+) -> torch.Tensor:
+    """
+    Computes the stability score for a batch of masks. The stability
+    score is the IoU between the binary masks obtained by thresholding
+    the predicted mask logits at high and low values.
+    """
+    # One mask is always contained inside the other.
+    # Save memory by preventing unnecessary cast to torch.int64
+    intersections = (
+        (masks > (mask_threshold + threshold_offset))
+        .sum(-1, dtype=torch.int16)
+        .sum(-1, dtype=torch.int32)
+    )
+    unions = (
+        (masks > (mask_threshold - threshold_offset))
+        .sum(-1, dtype=torch.int16)
+        .sum(-1, dtype=torch.int32)
+    )
+    return intersections / unions
+
+
+def build_point_grid(n_per_side: int) -> np.ndarray:
+    """Generates a 2D grid of points evenly spaced in [0,1]x[0,1]."""
+    offset = 1 / (2 * n_per_side)
+    points_one_side = np.linspace(offset, 1 - offset, n_per_side)
+    points_x = np.tile(points_one_side[None, :], (n_per_side, 1))
+    points_y = np.tile(points_one_side[:, None], (1, n_per_side))
+    points = np.stack([points_x, points_y], axis=-1).reshape(-1, 2)
+    return points
+
+
+def build_all_layer_point_grids(
+    n_per_side: int, n_layers: int, scale_per_layer: int
+) -> List[np.ndarray]:
+    """Generates point grids for all crop layers."""
+    points_by_layer = []
+    for i in range(n_layers + 1):
+        n_points = int(n_per_side / (scale_per_layer**i))
+        points_by_layer.append(build_point_grid(n_points))
+    return points_by_layer
+
+
+def generate_crop_boxes(
+    im_size: Tuple[int, ...], n_layers: int, overlap_ratio: float
+) -> Tuple[List[List[int]], List[int]]:
+    """
+    Generates a list of crop boxes of different sizes. Each layer
+    has (2**i)**2 boxes for the ith layer.
+    """
+    crop_boxes, layer_idxs = [], []
+    im_h, im_w = im_size
+    short_side = min(im_h, im_w)
+
+    # Original image
+    crop_boxes.append([0, 0, im_w, im_h])
+    layer_idxs.append(0)
+
+    def crop_len(orig_len, n_crops, overlap):
+        return int(math.ceil((overlap * (n_crops - 1) + orig_len) / n_crops))
+
+    for i_layer in range(n_layers):
+        n_crops_per_side = 2 ** (i_layer + 1)
+        overlap = int(overlap_ratio * short_side * (2 / n_crops_per_side))
+
+        crop_w = crop_len(im_w, n_crops_per_side, overlap)
+        crop_h = crop_len(im_h, n_crops_per_side, overlap)
+
+        crop_box_x0 = [int((crop_w - overlap) * i) for i in range(n_crops_per_side)]
+        crop_box_y0 = [int((crop_h - overlap) * i) for i in range(n_crops_per_side)]
+
+        # Crops in XYWH format
+        for x0, y0 in product(crop_box_x0, crop_box_y0):
+            box = [x0, y0, min(x0 + crop_w, im_w), min(y0 + crop_h, im_h)]
+            crop_boxes.append(box)
+            layer_idxs.append(i_layer + 1)
+
+    return crop_boxes, layer_idxs
+
+
+def uncrop_boxes_xyxy(boxes: torch.Tensor, crop_box: List[int]) -> torch.Tensor:
+    x0, y0, _, _ = crop_box
+    offset = torch.tensor([[x0, y0, x0, y0]], device=boxes.device)
+    # Check if boxes has a channel dimension
+    if len(boxes.shape) == 3:
+        offset = offset.unsqueeze(1)
+    return boxes + offset
+
+
+def uncrop_points(points: torch.Tensor, crop_box: List[int]) -> torch.Tensor:
+    x0, y0, _, _ = crop_box
+    offset = torch.tensor([[x0, y0]], device=points.device)
+    # Check if points has a channel dimension
+    if len(points.shape) == 3:
+        offset = offset.unsqueeze(1)
+    return points + offset
+
+
+def uncrop_masks(
+    masks: torch.Tensor, crop_box: List[int], orig_h: int, orig_w: int
+) -> torch.Tensor:
+    x0, y0, x1, y1 = crop_box
+    if x0 == 0 and y0 == 0 and x1 == orig_w and y1 == orig_h:
+        return masks
+    # Coordinate transform masks
+    pad_x, pad_y = orig_w - (x1 - x0), orig_h - (y1 - y0)
+    pad = (x0, pad_x - x0, y0, pad_y - y0)
+    return torch.nn.functional.pad(masks, pad, value=0)
+
+
+def remove_small_regions(
+    mask: np.ndarray, area_thresh: float, mode: str
+) -> Tuple[np.ndarray, bool]:
+    """
+    Removes small disconnected regions and holes in a mask. Returns the
+    mask and an indicator of if the mask has been modified.
+    """
+    import cv2  # type: ignore
+
+    assert mode in ["holes", "islands"]
+    correct_holes = mode == "holes"
+    working_mask = (correct_holes ^ mask).astype(np.uint8)
+    n_labels, regions, stats, _ = cv2.connectedComponentsWithStats(working_mask, 8)
+    sizes = stats[:, -1][1:]  # Row 0 is background label
+    small_regions = [i + 1 for i, s in enumerate(sizes) if s < area_thresh]
+    if len(small_regions) == 0:
+        return mask, False
+    fill_labels = [0] + small_regions
+    if not correct_holes:
+        fill_labels = [i for i in range(n_labels) if i not in fill_labels]
+        # If every region is below threshold, keep largest
+        if len(fill_labels) == 0:
+            fill_labels = [int(np.argmax(sizes)) + 1]
+    mask = np.isin(regions, fill_labels)
+    return mask, True
+
+
+def coco_encode_rle(uncompressed_rle: Dict[str, Any]) -> Dict[str, Any]:
+    from pycocotools import mask as mask_utils  # type: ignore
+
+    h, w = uncompressed_rle["size"]
+    rle = mask_utils.frPyObjects(uncompressed_rle, h, w)
+    rle["counts"] = rle["counts"].decode("utf-8")  # Necessary to serialize with json
+    return rle
+
+
+def batched_mask_to_box(masks: torch.Tensor) -> torch.Tensor:
+    """
+    Calculates boxes in XYXY format around masks. Return [0,0,0,0] for
+    an empty mask. For input shape C1xC2x...xHxW, the output shape is C1xC2x...x4.
+    """
+    # torch.max below raises an error on empty inputs, just skip in this case
+    if torch.numel(masks) == 0:
+        return torch.zeros(*masks.shape[:-2], 4, device=masks.device)
+
+    # Normalize shape to CxHxW
+    shape = masks.shape
+    h, w = shape[-2:]
+    if len(shape) > 2:
+        masks = masks.flatten(0, -3)
+    else:
+        masks = masks.unsqueeze(0)
+
+    # Get top and bottom edges
+    in_height, _ = torch.max(masks, dim=-1)
+    in_height_coords = in_height * torch.arange(h, device=in_height.device)[None, :]
+    bottom_edges, _ = torch.max(in_height_coords, dim=-1)
+    in_height_coords = in_height_coords + h * (~in_height)
+    top_edges, _ = torch.min(in_height_coords, dim=-1)
+
+    # Get left and right edges
+    in_width, _ = torch.max(masks, dim=-2)
+    in_width_coords = in_width * torch.arange(w, device=in_width.device)[None, :]
+    right_edges, _ = torch.max(in_width_coords, dim=-1)
+    in_width_coords = in_width_coords + w * (~in_width)
+    left_edges, _ = torch.min(in_width_coords, dim=-1)
+
+    # If the mask is empty the right edge will be to the left of the left edge.
+    # Replace these boxes with [0, 0, 0, 0]
+    empty_filter = (right_edges < left_edges) | (bottom_edges < top_edges)
+    out = torch.stack([left_edges, top_edges, right_edges, bottom_edges], dim=-1)
+    out = out * (~empty_filter).unsqueeze(-1)
+
+    # Return to original shape
+    if len(shape) > 2:
+        out = out.reshape(*shape[:-2], 4)
+    else:
+        out = out[0]
+
+    return out

segment_anything/utils/coreml.py

@@ -0,0 +1,64 @@
+import torch
+import torch.nn as nn
+from ..modeling import Sam
+from .amg import calculate_stability_score
+
+
+class SamCoreMLModel(nn.Module):
+    """
+    This model should not be called directly, but is used in CoreML export.
+    """
+
+    def __init__(
+        self,
+        model: Sam,
+        use_stability_score: bool = False
+    ) -> None:
+        super().__init__()
+        self.mask_decoder = model.mask_decoder
+        self.model = model
+        self.img_size = model.image_encoder.img_size
+        self.use_stability_score = use_stability_score
+        self.stability_score_offset = 1.0
+
+    def _embed_points(self, point_coords: torch.Tensor, point_labels: torch.Tensor) -> torch.Tensor:
+        point_coords = point_coords + 0.5
+        point_coords = point_coords / self.img_size
+        point_embedding = self.model.prompt_encoder.pe_layer._pe_encoding(point_coords)
+        point_labels = point_labels.unsqueeze(-1).expand_as(point_embedding)
+
+        point_embedding = point_embedding * (point_labels != -1)
+        point_embedding = point_embedding + self.model.prompt_encoder.not_a_point_embed.weight * (
+            point_labels == -1
+        )
+
+        for i in range(self.model.prompt_encoder.num_point_embeddings):
+            point_embedding = point_embedding + self.model.prompt_encoder.point_embeddings[
+                i
+            ].weight * (point_labels == i)
+
+        return point_embedding
+
+    @torch.no_grad()
+    def forward(
+        self,
+        image_embeddings: torch.Tensor,
+        point_coords: torch.Tensor,
+        point_labels: torch.Tensor,
+    ):
+        sparse_embedding = self._embed_points(point_coords, point_labels)
+        dense_embedding = self.model.prompt_encoder.no_mask_embed.weight.reshape(1, -1, 1, 1)
+
+        masks, scores = self.model.mask_decoder.predict_masks(
+            image_embeddings=image_embeddings,
+            image_pe=self.model.prompt_encoder.get_dense_pe(),
+            sparse_prompt_embeddings=sparse_embedding,
+            dense_prompt_embeddings=dense_embedding,
+        )
+
+        if self.use_stability_score:
+            scores = calculate_stability_score(
+                masks, self.model.mask_threshold, self.stability_score_offset
+            )
+
+        return scores, masks

segment_anything/utils/transforms.py

@@ -0,0 +1,102 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import numpy as np
+import torch
+from torch.nn import functional as F
+from torchvision.transforms.functional import resize, to_pil_image  # type: ignore
+
+from copy import deepcopy
+from typing import Tuple
+
+
+class ResizeLongestSide:
+    """
+    Resizes images to the longest side 'target_length', as well as provides
+    methods for resizing coordinates and boxes. Provides methods for
+    transforming both numpy array and batched torch tensors.
+    """
+
+    def __init__(self, target_length: int) -> None:
+        self.target_length = target_length
+
+    def apply_image(self, image: np.ndarray) -> np.ndarray:
+        """
+        Expects a numpy array with shape HxWxC in uint8 format.
+        """
+        target_size = self.get_preprocess_shape(image.shape[0], image.shape[1], self.target_length)
+        return np.array(resize(to_pil_image(image), target_size))
+
+    def apply_coords(self, coords: np.ndarray, original_size: Tuple[int, ...]) -> np.ndarray:
+        """
+        Expects a numpy array of length 2 in the final dimension. Requires the
+        original image size in (H, W) format.
+        """
+        old_h, old_w = original_size
+        new_h, new_w = self.get_preprocess_shape(
+            original_size[0], original_size[1], self.target_length
+        )
+        coords = deepcopy(coords).astype(float)
+        coords[..., 0] = coords[..., 0] * (new_w / old_w)
+        coords[..., 1] = coords[..., 1] * (new_h / old_h)
+        return coords
+
+    def apply_boxes(self, boxes: np.ndarray, original_size: Tuple[int, ...]) -> np.ndarray:
+        """
+        Expects a numpy array shape Bx4. Requires the original image size
+        in (H, W) format.
+        """
+        boxes = self.apply_coords(boxes.reshape(-1, 2, 2), original_size)
+        return boxes.reshape(-1, 4)
+
+    def apply_image_torch(self, image: torch.Tensor) -> torch.Tensor:
+        """
+        Expects batched images with shape BxCxHxW and float format. This
+        transformation may not exactly match apply_image. apply_image is
+        the transformation expected by the model.
+        """
+        # Expects an image in BCHW format. May not exactly match apply_image.
+        target_size = self.get_preprocess_shape(image.shape[2], image.shape[3], self.target_length)
+        return F.interpolate(
+            image, target_size, mode="bilinear", align_corners=False, antialias=True
+        )
+
+    def apply_coords_torch(
+        self, coords: torch.Tensor, original_size: Tuple[int, ...]
+    ) -> torch.Tensor:
+        """
+        Expects a torch tensor with length 2 in the last dimension. Requires the
+        original image size in (H, W) format.
+        """
+        old_h, old_w = original_size
+        new_h, new_w = self.get_preprocess_shape(
+            original_size[0], original_size[1], self.target_length
+        )
+        coords = deepcopy(coords).to(torch.float)
+        coords[..., 0] = coords[..., 0] * (new_w / old_w)
+        coords[..., 1] = coords[..., 1] * (new_h / old_h)
+        return coords
+
+    def apply_boxes_torch(
+        self, boxes: torch.Tensor, original_size: Tuple[int, ...]
+    ) -> torch.Tensor:
+        """
+        Expects a torch tensor with shape Bx4. Requires the original image
+        size in (H, W) format.
+        """
+        boxes = self.apply_coords_torch(boxes.reshape(-1, 2, 2), original_size)
+        return boxes.reshape(-1, 4)
+
+    @staticmethod
+    def get_preprocess_shape(oldh: int, oldw: int, long_side_length: int) -> Tuple[int, int]:
+        """
+        Compute the output size given input size and target long side length.
+        """
+        scale = long_side_length * 1.0 / max(oldh, oldw)
+        newh, neww = oldh * scale, oldw * scale
+        neww = int(neww + 0.5)
+        newh = int(newh + 0.5)
+        return (newh, neww)

utils/tools.py

@@ -0,0 +1,406 @@
+import os
+import sys
+
+import cv2
+import matplotlib.pyplot as plt
+import numpy as np
+import torch
+from PIL import Image
+
+
+def convert_box_xywh_to_xyxy(box):
+    x1 = box[0]
+    y1 = box[1]
+    x2 = box[0] + box[2]
+    y2 = box[1] + box[3]
+    return [x1, y1, x2, y2]
+
+
+def segment_image(image, bbox):
+    image_array = np.array(image)
+    segmented_image_array = np.zeros_like(image_array)
+    x1, y1, x2, y2 = bbox
+    segmented_image_array[y1:y2, x1:x2] = image_array[y1:y2, x1:x2]
+    segmented_image = Image.fromarray(segmented_image_array)
+    black_image = Image.new("RGB", image.size, (255, 255, 255))
+    # transparency_mask = np.zeros_like((), dtype=np.uint8)
+    transparency_mask = np.zeros(
+        (image_array.shape[0], image_array.shape[1]), dtype=np.uint8
+    )
+    transparency_mask[y1:y2, x1:x2] = 255
+    transparency_mask_image = Image.fromarray(transparency_mask, mode="L")
+    black_image.paste(segmented_image, mask=transparency_mask_image)
+    return black_image
+
+
+def format_results(masks, scores, logits, filter=0):
+    annotations = []
+    n = len(scores)
+    for i in range(n):
+        annotation = {}
+
+        mask = masks[i]
+        tmp = np.where(mask != 0)
+        if np.sum(mask) < filter:
+            continue
+        annotation["id"] = i
+        annotation["segmentation"] = mask
+        annotation["bbox"] = [
+            np.min(tmp[0]),
+            np.min(tmp[1]),
+            np.max(tmp[1]),
+            np.max(tmp[0]),
+        ]
+        annotation["score"] = scores[i]
+        annotation["area"] = annotation["segmentation"].sum()
+        annotations.append(annotation)
+    return annotations
+
+
+def filter_masks(annotations):  # filter the overlap mask
+    annotations.sort(key=lambda x: x["area"], reverse=True)
+    to_remove = set()
+    for i in range(0, len(annotations)):
+        a = annotations[i]
+        for j in range(i + 1, len(annotations)):
+            b = annotations[j]
+            if i != j and j not in to_remove:
+                # check if
+                if b["area"] < a["area"]:
+                    if (a["segmentation"] & b["segmentation"]).sum() / b[
+                        "segmentation"
+                    ].sum() > 0.8:
+                        to_remove.add(j)
+
+    return [a for i, a in enumerate(annotations) if i not in to_remove], to_remove
+
+
+def get_bbox_from_mask(mask):
+    mask = mask.astype(np.uint8)
+    contours, hierarchy = cv2.findContours(
+        mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE
+    )
+    x1, y1, w, h = cv2.boundingRect(contours[0])
+    x2, y2 = x1 + w, y1 + h
+    if len(contours) > 1:
+        for b in contours:
+            x_t, y_t, w_t, h_t = cv2.boundingRect(b)
+            # 将多个bbox合并成一个
+            x1 = min(x1, x_t)
+            y1 = min(y1, y_t)
+            x2 = max(x2, x_t + w_t)
+            y2 = max(y2, y_t + h_t)
+        h = y2 - y1
+        w = x2 - x1
+    return [x1, y1, x2, y2]
+
+
+def fast_process(
+    annotations, args, mask_random_color, bbox=None, points=None, edges=False
+):
+    if isinstance(annotations[0], dict):
+        annotations = [annotation["segmentation"] for annotation in annotations]
+    result_name = os.path.basename(args.img_path)
+    image = cv2.imread(args.img_path)
+    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+    original_h = image.shape[0]
+    original_w = image.shape[1]
+    if sys.platform == "darwin":
+        plt.switch_backend("TkAgg")
+    plt.figure(figsize=(original_w / 100, original_h / 100))
+    # Add subplot with no margin.
+    plt.subplots_adjust(top=1, bottom=0, right=1, left=0, hspace=0, wspace=0)
+    plt.margins(0, 0)
+    plt.gca().xaxis.set_major_locator(plt.NullLocator())
+    plt.gca().yaxis.set_major_locator(plt.NullLocator())
+    plt.imshow(image)
+    if args.better_quality == True:
+        if isinstance(annotations[0], torch.Tensor):
+            annotations = np.array(annotations.cpu())
+        for i, mask in enumerate(annotations):
+            mask = cv2.morphologyEx(
+                mask.astype(np.uint8), cv2.MORPH_CLOSE, np.ones((3, 3), np.uint8)
+            )
+            annotations[i] = cv2.morphologyEx(
+                mask.astype(np.uint8), cv2.MORPH_OPEN, np.ones((8, 8), np.uint8)
+            )
+    if args.device == "cpu":
+        annotations = np.array(annotations)
+        fast_show_mask(
+            annotations,
+            plt.gca(),
+            random_color=mask_random_color,
+            bbox=bbox,
+            points=points,
+            point_label=args.point_label,
+            retinamask=args.retina,
+            target_height=original_h,
+            target_width=original_w,
+        )
+    else:
+        if isinstance(annotations[0], np.ndarray):
+            annotations = torch.from_numpy(annotations)
+        fast_show_mask_gpu(
+            annotations,
+            plt.gca(),
+            random_color=args.randomcolor,
+            bbox=bbox,
+            points=points,
+            point_label=args.point_label,
+            retinamask=args.retina,
+            target_height=original_h,
+            target_width=original_w,
+        )
+    if isinstance(annotations, torch.Tensor):
+        annotations = annotations.cpu().numpy()
+    if args.withContours == True:
+        contour_all = []
+        temp = np.zeros((original_h, original_w, 1))
+        for i, mask in enumerate(annotations):
+            if type(mask) == dict:
+                mask = mask["segmentation"]
+            annotation = mask.astype(np.uint8)
+            if args.retina == False:
+                annotation = cv2.resize(
+                    annotation,
+                    (original_w, original_h),
+                    interpolation=cv2.INTER_NEAREST,
+                )
+            contours, hierarchy = cv2.findContours(
+                annotation, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE
+            )
+            for contour in contours:
+                contour_all.append(contour)
+        cv2.drawContours(temp, contour_all, -1, (255, 255, 255), 2)
+        color = np.array([0 / 255, 0 / 255, 255 / 255, 0.8])
+        contour_mask = temp / 255 * color.reshape(1, 1, -1)
+        plt.imshow(contour_mask)
+
+    save_path = args.output
+    if not os.path.exists(save_path):
+        os.makedirs(save_path)
+    plt.axis("off")
+    fig = plt.gcf()
+    plt.draw()
+
+    try:
+        buf = fig.canvas.tostring_rgb()
+    except AttributeError:
+        fig.canvas.draw()
+        buf = fig.canvas.tostring_rgb()
+
+    cols, rows = fig.canvas.get_width_height()
+    img_array = np.fromstring(buf, dtype=np.uint8).reshape(rows, cols, 3)
+    cv2.imwrite(
+        os.path.join(save_path, result_name), cv2.cvtColor(img_array, cv2.COLOR_RGB2BGR)
+    )
+
+
+# CPU post process
+def fast_show_mask(
+    annotation,
+    ax,
+    random_color=False,
+    bbox=None,
+    points=None,
+    point_label=None,
+    retinamask=True,
+    target_height=960,
+    target_width=960,
+):
+    msak_sum = annotation.shape[0]
+    height = annotation.shape[1]
+    weight = annotation.shape[2]
+    # 将annotation 按照面积 排序
+    areas = np.sum(annotation, axis=(1, 2))
+    sorted_indices = np.argsort(areas)
+    annotation = annotation[sorted_indices]
+
+    index = (annotation != 0).argmax(axis=0)
+    if random_color == True:
+        color = np.random.random((msak_sum, 1, 1, 3))
+    else:
+        color = np.ones((msak_sum, 1, 1, 3)) * np.array(
+            [30 / 255, 144 / 255, 255 / 255]
+        )
+    transparency = np.ones((msak_sum, 1, 1, 1)) * 0.6
+    visual = np.concatenate([color, transparency], axis=-1)
+    mask_image = np.expand_dims(annotation, -1) * visual
+
+    show = np.zeros((height, weight, 4))
+    h_indices, w_indices = np.meshgrid(
+        np.arange(height), np.arange(weight), indexing="ij"
+    )
+    indices = (index[h_indices, w_indices], h_indices, w_indices, slice(None))
+    # 使用向量化索引更新show的值
+    show[h_indices, w_indices, :] = mask_image[indices]
+    if bbox is not None:
+        x1, y1, x2, y2 = bbox
+        ax.add_patch(
+            plt.Rectangle(
+                (x1, y1), x2 - x1, y2 - y1, fill=False, edgecolor="b", linewidth=1
+            )
+        )
+    # draw point
+    if points is not None:
+        plt.scatter(
+            [point[0] for i, point in enumerate(points) if point_label[i] == 1],
+            [point[1] for i, point in enumerate(points) if point_label[i] == 1],
+            s=20,
+            c="y",
+        )
+        plt.scatter(
+            [point[0] for i, point in enumerate(points) if point_label[i] == 0],
+            [point[1] for i, point in enumerate(points) if point_label[i] == 0],
+            s=20,
+            c="m",
+        )
+
+    if retinamask == False:
+        show = cv2.resize(
+            show, (target_width, target_height), interpolation=cv2.INTER_NEAREST
+        )
+    ax.imshow(show)
+
+
+def fast_show_mask_gpu(
+    annotation,
+    ax,
+    random_color=False,
+    bbox=None,
+    points=None,
+    point_label=None,
+    retinamask=True,
+    target_height=960,
+    target_width=960,
+):
+    msak_sum = annotation.shape[0]
+    height = annotation.shape[1]
+    weight = annotation.shape[2]
+    areas = torch.sum(annotation, dim=(1, 2))
+    sorted_indices = torch.argsort(areas, descending=False)
+    annotation = annotation[sorted_indices]
+    # 找每个位置第一个非零值下标
+    index = (annotation != 0).to(torch.long).argmax(dim=0)
+    if random_color == True:
+        color = torch.rand((msak_sum, 1, 1, 3)).to(annotation.device)
+    else:
+        color = torch.ones((msak_sum, 1, 1, 3)).to(annotation.device) * torch.tensor(
+            [30 / 255, 144 / 255, 255 / 255]
+        ).to(annotation.device)
+    transparency = torch.ones((msak_sum, 1, 1, 1)).to(annotation.device) * 0.6
+    visual = torch.cat([color, transparency], dim=-1)
+    mask_image = torch.unsqueeze(annotation, -1) * visual
+    # 按index取数，index指每个位置选哪个batch的数，把mask_image转成一个batch的形式
+    show = torch.zeros((height, weight, 4)).to(annotation.device)
+    h_indices, w_indices = torch.meshgrid(
+        torch.arange(height), torch.arange(weight), indexing="ij"
+    )
+    indices = (index[h_indices, w_indices], h_indices, w_indices, slice(None))
+    # 使用向量化索引更新show的值
+    show[h_indices, w_indices, :] = mask_image[indices]
+    show_cpu = show.cpu().numpy()
+    if bbox is not None:
+        x1, y1, x2, y2 = bbox
+        ax.add_patch(
+            plt.Rectangle(
+                (x1, y1), x2 - x1, y2 - y1, fill=False, edgecolor="b", linewidth=1
+            )
+        )
+    # draw point
+    if points is not None:
+        plt.scatter(
+            [point[0] for i, point in enumerate(points) if point_label[i] == 1],
+            [point[1] for i, point in enumerate(points) if point_label[i] == 1],
+            s=20,
+            c="y",
+        )
+        plt.scatter(
+            [point[0] for i, point in enumerate(points) if point_label[i] == 0],
+            [point[1] for i, point in enumerate(points) if point_label[i] == 0],
+            s=20,
+            c="m",
+        )
+    if retinamask == False:
+        show_cpu = cv2.resize(
+            show_cpu, (target_width, target_height), interpolation=cv2.INTER_NEAREST
+        )
+    ax.imshow(show_cpu)
+
+
+def crop_image(annotations, image_like):
+    if isinstance(image_like, str):
+        image = Image.open(image_like)
+    else:
+        image = image_like
+    ori_w, ori_h = image.size
+    mask_h, mask_w = annotations[0]["segmentation"].shape
+    if ori_w != mask_w or ori_h != mask_h:
+        image = image.resize((mask_w, mask_h))
+    cropped_boxes = []
+    cropped_images = []
+    not_crop = []
+    filter_id = []
+    # annotations, _ = filter_masks(annotations)
+    # filter_id = list(_)
+    for _, mask in enumerate(annotations):
+        if np.sum(mask["segmentation"]) <= 100:
+            filter_id.append(_)
+            continue
+        bbox = get_bbox_from_mask(mask["segmentation"])  # mask 的 bbox
+        cropped_boxes.append(segment_image(image, bbox))  # 保存裁剪的图片
+        # cropped_boxes.append(segment_image(image,mask["segmentation"]))
+        cropped_images.append(bbox)  # 保存裁剪的图片的bbox
+
+    return cropped_boxes, cropped_images, not_crop, filter_id, annotations
+
+
+def box_prompt(masks, bbox, target_height, target_width):
+    h = masks.shape[1]
+    w = masks.shape[2]
+    if h != target_height or w != target_width:
+        bbox = [
+            int(bbox[0] * w / target_width),
+            int(bbox[1] * h / target_height),
+            int(bbox[2] * w / target_width),
+            int(bbox[3] * h / target_height),
+        ]
+    bbox[0] = round(bbox[0]) if round(bbox[0]) > 0 else 0
+    bbox[1] = round(bbox[1]) if round(bbox[1]) > 0 else 0
+    bbox[2] = round(bbox[2]) if round(bbox[2]) < w else w
+    bbox[3] = round(bbox[3]) if round(bbox[3]) < h else h
+
+    # IoUs = torch.zeros(len(masks), dtype=torch.float32)
+    bbox_area = (bbox[3] - bbox[1]) * (bbox[2] - bbox[0])
+
+    masks_area = torch.sum(masks[:, bbox[1] : bbox[3], bbox[0] : bbox[2]], dim=(1, 2))
+    orig_masks_area = torch.sum(masks, dim=(1, 2))
+
+    union = bbox_area + orig_masks_area - masks_area
+    IoUs = masks_area / union
+    max_iou_index = torch.argmax(IoUs)
+
+    return masks[max_iou_index].cpu().numpy(), max_iou_index
+
+
+def point_prompt(masks, points, point_label, target_height, target_width):  # numpy 处理
+    h = masks[0]["segmentation"].shape[0]
+    w = masks[0]["segmentation"].shape[1]
+    if h != target_height or w != target_width:
+        points = [
+            [int(point[0] * w / target_width), int(point[1] * h / target_height)]
+            for point in points
+        ]
+    onemask = np.zeros((h, w))
+    for i, annotation in enumerate(masks):
+        if type(annotation) == dict:
+            mask = annotation["segmentation"]
+        else:
+            mask = annotation
+        for i, point in enumerate(points):
+            if mask[point[1], point[0]] == 1 and point_label[i] == 1:
+                onemask += mask
+            if mask[point[1], point[0]] == 1 and point_label[i] == 0:
+                onemask -= mask
+    onemask = onemask >= 1
+    return onemask, 0

utils/tools_gradio.py

@@ -0,0 +1,192 @@
+import cv2
+import matplotlib.pyplot as plt
+import numpy as np
+import torch
+from PIL import Image
+
+
+def fast_process(
+    annotations,
+    image,
+    device,
+    scale,
+    better_quality=False,
+    mask_random_color=True,
+    bbox=None,
+    use_retina=True,
+    withContours=True,
+):
+    if isinstance(annotations[0], dict):
+        annotations = [annotation["segmentation"] for annotation in annotations]
+
+    original_h = image.height
+    original_w = image.width
+    if better_quality:
+        if isinstance(annotations[0], torch.Tensor):
+            annotations = np.array(annotations.cpu())
+        for i, mask in enumerate(annotations):
+            mask = cv2.morphologyEx(
+                mask.astype(np.uint8), cv2.MORPH_CLOSE, np.ones((3, 3), np.uint8)
+            )
+            annotations[i] = cv2.morphologyEx(
+                mask.astype(np.uint8), cv2.MORPH_OPEN, np.ones((8, 8), np.uint8)
+            )
+    if device == "cpu":
+        annotations = np.array(annotations)
+        inner_mask = fast_show_mask(
+            annotations,
+            plt.gca(),
+            random_color=mask_random_color,
+            bbox=bbox,
+            retinamask=use_retina,
+            target_height=original_h,
+            target_width=original_w,
+        )
+    else:
+        if isinstance(annotations[0], np.ndarray):
+            annotations = np.array(annotations)
+            annotations = torch.from_numpy(annotations)
+        inner_mask = fast_show_mask_gpu(
+            annotations,
+            plt.gca(),
+            random_color=mask_random_color,
+            bbox=bbox,
+            retinamask=use_retina,
+            target_height=original_h,
+            target_width=original_w,
+        )
+    if isinstance(annotations, torch.Tensor):
+        annotations = annotations.cpu().numpy()
+
+    if withContours:
+        contour_all = []
+        temp = np.zeros((original_h, original_w, 1))
+        for i, mask in enumerate(annotations):
+            if type(mask) == dict:
+                mask = mask["segmentation"]
+            annotation = mask.astype(np.uint8)
+            if use_retina == False:
+                annotation = cv2.resize(
+                    annotation,
+                    (original_w, original_h),
+                    interpolation=cv2.INTER_NEAREST,
+                )
+            contours, _ = cv2.findContours(
+                annotation, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE
+            )
+            for contour in contours:
+                contour_all.append(contour)
+        cv2.drawContours(temp, contour_all, -1, (255, 255, 255), 2 // scale)
+        color = np.array([0 / 255, 0 / 255, 255 / 255, 0.9])
+        contour_mask = temp / 255 * color.reshape(1, 1, -1)
+
+    image = image.convert("RGBA")
+    overlay_inner = Image.fromarray((inner_mask * 255).astype(np.uint8), "RGBA")
+    image.paste(overlay_inner, (0, 0), overlay_inner)
+
+    if withContours:
+        overlay_contour = Image.fromarray((contour_mask * 255).astype(np.uint8), "RGBA")
+        image.paste(overlay_contour, (0, 0), overlay_contour)
+
+    return image
+
+
+# CPU post process
+def fast_show_mask(
+    annotation,
+    ax,
+    random_color=False,
+    bbox=None,
+    retinamask=True,
+    target_height=960,
+    target_width=960,
+):
+    mask_sum = annotation.shape[0]
+    height = annotation.shape[1]
+    weight = annotation.shape[2]
+    # 将annotation 按照面积 排序
+    areas = np.sum(annotation, axis=(1, 2))
+    sorted_indices = np.argsort(areas)[::1]
+    annotation = annotation[sorted_indices]
+
+    index = (annotation != 0).argmax(axis=0)
+    if random_color == True:
+        color = np.random.random((mask_sum, 1, 1, 3))
+    else:
+        color = np.ones((mask_sum, 1, 1, 3)) * np.array(
+            [30 / 255, 144 / 255, 255 / 255]
+        )
+    transparency = np.ones((mask_sum, 1, 1, 1)) * 0.6
+    visual = np.concatenate([color, transparency], axis=-1)
+    mask_image = np.expand_dims(annotation, -1) * visual
+
+    mask = np.zeros((height, weight, 4))
+
+    h_indices, w_indices = np.meshgrid(
+        np.arange(height), np.arange(weight), indexing="ij"
+    )
+    indices = (index[h_indices, w_indices], h_indices, w_indices, slice(None))
+
+    mask[h_indices, w_indices, :] = mask_image[indices]
+    if bbox is not None:
+        x1, y1, x2, y2 = bbox
+        ax.add_patch(
+            plt.Rectangle(
+                (x1, y1), x2 - x1, y2 - y1, fill=False, edgecolor="b", linewidth=1
+            )
+        )
+
+    if retinamask == False:
+        mask = cv2.resize(
+            mask, (target_width, target_height), interpolation=cv2.INTER_NEAREST
+        )
+
+    return mask
+
+
+def fast_show_mask_gpu(
+    annotation,
+    ax,
+    random_color=False,
+    bbox=None,
+    retinamask=True,
+    target_height=960,
+    target_width=960,
+):
+    device = annotation.device
+    mask_sum = annotation.shape[0]
+    height = annotation.shape[1]
+    weight = annotation.shape[2]
+    areas = torch.sum(annotation, dim=(1, 2))
+    sorted_indices = torch.argsort(areas, descending=False)
+    annotation = annotation[sorted_indices]
+    # 找每个位置第一个非零值下标
+    index = (annotation != 0).to(torch.long).argmax(dim=0)
+    if random_color == True:
+        color = torch.rand((mask_sum, 1, 1, 3)).to(device)
+    else:
+        color = torch.ones((mask_sum, 1, 1, 3)).to(device) * torch.tensor(
+            [30 / 255, 144 / 255, 255 / 255]
+        ).to(device)
+    transparency = torch.ones((mask_sum, 1, 1, 1)).to(device) * 0.6
+    visual = torch.cat([color, transparency], dim=-1)
+    mask_image = torch.unsqueeze(annotation, -1) * visual
+    # 按index取数，index指每个位置选哪个batch的数，把mask_image转成一个batch的形式
+    mask = torch.zeros((height, weight, 4)).to(device)
+    h_indices, w_indices = torch.meshgrid(torch.arange(height), torch.arange(weight))
+    indices = (index[h_indices, w_indices], h_indices, w_indices, slice(None))
+    # 使用向量化索引更新show的值
+    mask[h_indices, w_indices, :] = mask_image[indices]
+    mask_cpu = mask.cpu().numpy()
+    if bbox is not None:
+        x1, y1, x2, y2 = bbox
+        ax.add_patch(
+            plt.Rectangle(
+                (x1, y1), x2 - x1, y2 - y1, fill=False, edgecolor="b", linewidth=1
+            )
+        )
+    if retinamask == False:
+        mask_cpu = cv2.resize(
+            mask_cpu, (target_width, target_height), interpolation=cv2.INTER_NEAREST
+        )
+    return mask_cpu

weights/edge_sam_3x.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a3e3261da9e2f4c8154410885689ae57d09cbf4cf5d0c848b9d80b6df336ece
+size 38807540