English

Sub-pixel face landmarks using heatmaps and a bag of tricks

Computer Vision and Pattern Recognition 2021-03-09 v2 Artificial Intelligence Machine Learning

Abstract

Accurate face landmark localization is an essential part of face recognition, reconstruction and morphing. To accurately localize face landmarks, we present our heatmap regression approach. Each model consists of a MobileNetV2 backbone followed by several upscaling layers, with different tricks to optimize both performance and inference cost. We use five na\"ive face landmarks from a publicly available face detector to position and align the face instead of using the bounding box like traditional methods. Moreover, we show by adding random rotation, displacement and scaling -- after alignment -- that the model is more sensitive to the face position than orientation. We also show that it is possible to reduce the upscaling complexity by using a mixture of deconvolution and pixel-shuffle layers without impeding localization performance. We present our state-of-the-art face landmark localization model (ranking second on The 2nd Grand Challenge of 106-Point Facial Landmark Localization validation set). Finally, we test the effect on face recognition using these landmarks, using a publicly available model and benchmarks.

Keywords

Cite

@article{arxiv.2103.03059,
  title  = {Sub-pixel face landmarks using heatmaps and a bag of tricks},
  author = {Samuel W. F. Earp and Aubin Samacoits and Sanjana Jain and Pavit Noinongyao and Siwa Boonpunmongkol},
  journal= {arXiv preprint arXiv:2103.03059},
  year   = {2021}
}

Comments

10 pages, 1 figure

R2 v1 2026-06-23T23:45:15.319Z