raja-kumar/dense-face-alignment

dense face landmark detection

dense-face-alignment

Most of the existing face-alignments detects the standard 68 face landmarks. However, there is a need of dense facial landmarks for tasks such 3d face reconstruction, face recognition etc. DAD-3DHeads proposed a dense 2d landmark detector based on 3D face reconstruction. However, their performance deteriorates when face is only a part of the image. see figure below. To improve on this, I used mediapipe facial detector to detect the face, use the detected face to DADNet and then project back the detected landmarks in the original image coordinate.

Dependencies

pip install -r requirements.txt

Note: if you are using mac silicon replace mediapipe with mediapipe-silicon in requirements.txt

How to replicate the results

python run.py -h

sample command

python run.py --data_path sample_input --num_points 445

sample command for custom points

python run.py --data_path sample_input --custom_index model_training/model/static/face_keypoints/ids1.npy

Citation

If you found this work useful for your research, please consider citing our paper

@inproceedings{kumar2023disjoint,
  title={Disjoint Pose and Shape for 3D Face Reconstruction},
  author={Kumar, Raja and Luo, Jiahao and Pang, Alex and Davis, James},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops},
  pages={3115--3125},
  year={2023}
}

Attributions

this repo is built on top of base code from DAD-3DHeads

dense-face-alignment's Issues

cuda error

RuntimeError: CUDA error: no kernel image is available for execution on the device