PropagationNet: Propagate Points to Curve to Learn Structure Information

TL;DR

PropagationNet introduces a structure-infused face alignment method using heatmap propagation and a novel loss function, significantly improving accuracy in challenging unconstrained scenarios.

Contribution

The paper proposes a novel face alignment algorithm with heatmap propagation and Focal Wing Loss, addressing structure modeling and difficult sample mining.

Findings

Outperforms state-of-the-art on WFLW, 300W, and COFW benchmarks.

Achieves 4.05% mean error on WFLW, 2.93% on 300W, and 3.71% on COFW.

Effectively handles large head poses, expressions, and illumination variations.

Abstract

Deep learning technique has dramatically boosted the performance of face alignment algorithms. However, due to large variability and lack of samples, the alignment problem in unconstrained situations, \emph{e.g}\onedot large head poses, exaggerated expression, and uneven illumination, is still largely unsolved. In this paper, we explore the instincts and reasons behind our two proposals, \emph{i.e}\onedot Propagation Module and Focal Wing Loss, to tackle the problem. Concretely, we present a novel structure-infused face alignment algorithm based on heatmap regression via propagating landmark heatmaps to boundary heatmaps, which provide structure information for further attention map generation. Moreover, we propose a Focal Wing Loss for mining and emphasizing the difficult samples under in-the-wild condition. In addition, we adopt methods like CoordConv and Anti-aliased CNN from other fields that address the shift-variance problem of CNN for face alignment. When implementing extensive experiments on different benchmarks, \emph{i.e}\onedot WFLW, 300W, and COFW, our method outperforms state-of-the-arts by a significant margin. Our proposed approach achieves 4.05\% mean error on WFLW, 2.93\% mean error on 300W full-set, and 3.71\% mean error on COFW.