Skip to Main content Skip to Navigation
Journal articles

2D Wasserstein Loss for Robust Facial Landmark Detection

Abstract : The recent performance of facial landmark detection has been significantly improved by using deep Convolutional Neural Networks (CNNs), especially the Heatmap Regression Models (HRMs). Although their performance on common benchmark datasets has reached a high level, the robustness of these models still remains a challenging problem in the practical use under noisy conditions of realistic environments. Contrary to most existing work focusing on the design of new models, we argue that improving the robustness requires rethinking many other aspects, including the use of datasets, the format of landmark annotation, the evaluation metric as well as the training and detection algorithm itself. In this paper, we propose a novel method for robust facial landmark detection, using a loss function based on the 2D Wasserstein distance combined with a new landmark coordinate sampling relying on the barycenter of the individual probability distributions. Our method can be plugged-and-play on most state-of-the-art HRMs with neither additional complexity nor structural modifications of the models. Further, with the large performance increase, we found that current evaluation metrics can no longer fully reflect the robustness of these models. Therefore, we propose several improvements to the standard evaluation protocol. Extensive experimental results on both traditional evaluation metrics and our evaluation metrics demonstrate that our approach significantly improves the robustness of state-of-the-art facial landmark detection models.
Document type :
Journal articles
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03162012
Contributor : Yongzhe Yan Connect in order to contact the contributor
Submitted on : Monday, March 8, 2021 - 11:23:37 AM
Last modification on : Tuesday, January 4, 2022 - 6:47:28 AM
Long-term archiving on: : Wednesday, June 9, 2021 - 6:47:31 PM

File

RevisionV2.pdf
Files produced by the author(s)

Identifiers

Citation

Yongzhe Yan, Stefan Duffner, Priyanka Phutane, Anthony Berthelier, Christophe Blanc, et al.. 2D Wasserstein Loss for Robust Facial Landmark Detection. Pattern Recognition, Elsevier, 2021, ⟨10.1016/j.patcog.2021.107945⟩. ⟨hal-03162012⟩

Share

Metrics

Record views

88

Files downloads

82