abhi1kumar/luvli

[CVPR 2020] Re-hosting of the LUVLi Face Alignment codebase. Please download the codebase from the original MERL website by agreeing to all terms and conditions. By using this code, you agree to MERL's research-only licensing terms.

LUVLi and UGLLI Face Alignment

The code is officially available here

LUVLi Face Alignment: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihood, CVPR 2020

[slides], [1min_talk], [supp],[demo]

UGLLI Face Alignment: Estimating Uncertainty with Gaussian Log-Likelihood Loss, ICCV Workshops on Statistical Deep Learning in Computer Vision 2019

[slides], [poster], [news], [Best Oral Presentation Award]

References

Please cite the following papers if you find this repository useful:

@inproceedings{kumar2020luvli,
  title={LUVLi Face Alignment: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihood},
  author={Kumar, Abhinav and Marks, Tim K. and Mou, Wenxuan and Wang, Ye and Jones, Michael and Cherian, Anoop and Koike-Akino, Toshiaki and Liu, Xiaoming and Feng, Chen},
  booktitle={IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2020}
}

@inproceedings{kumar2019uglli,
  title={UGLLI Face Alignment: Estimating Uncertainty with Gaussian Log-Likelihood Loss},
  author={Kumar, Abhinav and Marks, Tim K and Mou, Wenxuan and Feng, Chen and Liu, Xiaoming},
  booktitle={ICCV Workshops on Statistical Deep Learning in Computer Vision},
  year={2019}
}

Evaluation of our pre-trained models

Split	Name	Directory	LUVLi	UGLLI
1	300-W Split 1	run_108	lr-0.00002-49.pth.tar	-
2	300-W Split 2	run_109	lr-0.00002-49.pth.tar	-
3	AFLW-19	run_507	lr-0.00002-49.pth.tar	-
4	WFLW	run_1005	lr-0.00002-49.pth.tar	-
5	MERL-RAV (AFLW_ours)	run_5004	lr-0.00002-49.pth.tar	-
1	300-W Split 1	run_924	-	lr-0.00002-39.pth.tar
2	300-W Split 2	run_940	-	lr-0.00002-39.pth.tar

Copy the pre-trained models to the abhinav_model_dir first. The directory structure should look like this:

./FaceAlignmentUncertainty/
|--- abhinav_model_dir/
|           |--- run_108
|           |       |--lr-0.00002-49.pth.tar
|           |
|           |--- run_109
|           |       |--lr-0.00002-49.pth.tar
|           |
|           |--- run_507
|           |       |--lr-0.00002-49.pth.tar
|           |
|           |--- run_1005
|           |       |--lr-0.00002-49.pth.tar
|           |
|           |--- run_5004
|           |       |--lr-0.00002-49.pth.tar
|  ...

Next type the following:

./scripts_evaluation.sh

In case you want to get our qualitative plots and also the transformed figures, type:

python plot/show_300W_images_overlaid_with_uncertainties.py --exp_id abhinav_model_dir/run_109_evaluate/ --laplacian
python plot/plot_uncertainties_in_transformed_space.py          -i run_109_evaluate/300W_test --laplacian
python plot/plot_residual_covariance_vs_predicted_covariance.py -i run_109_evaluate --laplacian
python plot/plot_histogram_smallest_eigen_value.py              -i run_109_evaluate --laplacian

luvli's Issues

Pretrained models

Hi, will you share your inference code and trained weights file? I'd like to try on some my own images. Thanks!

The params and flops

Hi, A nice work in face alignment!
For better compare to other works, I wanna to know the params and flops (in WFLW,98pts) on LUVLI.

Running Inference

Hi, thanks for the great work. I've downloaded the zip file with the codes and tried running but there seem to be no clear cut indication on how to run so can you please state the requirements for running the inference. I'm choosing to run this code as i need the visibility values for each landmark. Any help regarding running the inference would be highly helpful.