|
--- |
|
license: agpl-3.0 |
|
library: ultralytics |
|
tags: |
|
- object-detection |
|
- pytorch |
|
- roboflow-universe |
|
- pickle |
|
- face-detection |
|
--- |
|
# Face Detection using YOLOv8 |
|
|
|
This model was fine tuned on a dataset of over 10k images containing human faces. The model was fine tuned for 100 epochs with a batch size of 16 on a single NVIDIA V100 16GB GPU, it took around 140 minutes for the fine tuning to complete. |
|
|
|
## Downstream Tasks |
|
|
|
- __Face Detection__: This model can directly use this model for face detection or it can be further fine tuned own a custom dataset to improve the prediction capabilities. |
|
- __Face Recognition__: This model can be fine tuned to for face recognition tasks as well, create a dataset with the images of faces and label them accordingly using name or any ID and then use this model as a base model for fine tuning. |
|
|
|
# Example Usage |
|
|
|
```python |
|
# load libraries |
|
from huggingface_hub import hf_hub_download |
|
from ultralytics import YOLO |
|
from supervision import Detections |
|
from PIL import Image |
|
|
|
# download model |
|
model_path = hf_hub_download(repo_id="arnabdhar/YOLOv8-Face-Detection", filename="model.pt") |
|
|
|
# load model |
|
model = YOLO(model_path) |
|
|
|
# inference |
|
image_path = "/path/to/image" |
|
output = model(Image.open(image_path)) |
|
results = Detections.from_ultralytics(output[0]) |
|
|
|
``` |
|
|
|
|
|
# Links |
|
|
|
- __Dataset Source__: [Roboflow Universe](https://universe.roboflow.com/large-benchmark-datasets/wider-face-ndtcz/dataset/1) |
|
- __Weights & Biases__: [Run Details](https://wandb.ai/2wb2ndur/Face-Detection/overview?workspace=user-2wb2ndur) |
|
- __Training Artifacts__: [training-artifacts](./fine-tune-artifacts/) |
