add README

README.md

@@ -0,0 +1,19 @@
+## Data Format
+
+Here we explain the `poses_bounds.npy` file format. This file stores a numpy array of size Nx17 (where N is the number of input videos). You can load the data using the following codes.
+
+```
+poses_arr = np.load(os.path.join(basedir, 'poses_bounds.npy'))
+poses = poses_arr[:, :-2].reshape([-1, 3, 5]).transpose([1,2,0])
+bds = poses_arr[:, -2:].transpose([1,0])
+```
+
+Each row of length 17 gets reshaped into a 3x5 pose matrix and 2 depth values that bound the closest and farthest scene content from that point of view.
+
+The pose matrix is a 3x4 camera-to-world affine transform concatenated with a 3x1 column `[image height, image width, focal length]` to represent the intrinsics (we assume the principal point is centered and that the focal length is the same for both x and y). 
+
+<big>NOTE: In our dataset, the focal length for different cameras are different!!!</big>
+
+The right-handed coordinate system of the the rotation (first 3x3 block in the camera-to-world transform) is as follows: from the point of view of the camera, the three axes are `[down, right, backwards]` which some people might consider to be `[-y,x,z]`, where the camera is looking along `-z`. (The more conventional frame `[x,y,z]` is `[right, up, backwards]`. The COLMAP frame is `[right, down, forwards]` or `[x,-y,-z]`.)
+
+We also provide an example of our dataloader in `dataloader.py`.

dataloader.py

@@ -0,0 +1,43 @@
+import torch
+from torch.utils.data import Dataset
+import glob
+import numpy as np
+import os
+from tqdm import tqdm
+
+
+
+class Robo360(Dataset):
+    def __init__(self, datadir, downsample=4):
+
+        self.root_dir = datadir
+        self.downsample = downsample
+
+        self.read_meta()
+
+
+    def read_meta(self):
+
+        poses_bounds = np.load(os.path.join(self.root_dir, 'poses_bounds.npy'))  # (N_cams, 17)
+
+        poses = poses_bounds[:, :15].reshape(-1, 3, 5)  # (N_images, 3, 5)
+        self.near_fars = poses_bounds[:, -2:]  # (N_images, 2)
+
+        # Step 1: rescale focal length according to training resolution
+        H, W, _ = poses[0, :, -1]
+        self.focal = poses[:, -1, -1]
+        self.img_wh = np.array([int(W / self.downsample), int(H / self.downsample)])
+        self.focal = self.focal * self.img_wh[0] / W
+
+        # Step 2: correct poses
+        # Original poses has rotation in form "down right back", change to "right up back"
+        # See https://github.com/bmild/nerf/issues/34
+        self.poses = np.concatenate([poses[..., 1:2], -poses[..., :1], poses[..., 2:4]], -1)
+
+
+
+    def __len__(self):
+        return 0
+
+    def __getitem__(self, idx):
+        return None