Robust face alignment by dual-attentional spatial-aware capsule networks

作者：

Highlights：

• We propose an hourglass capsule network to build a more accurate facial inter-feature spatial constrained model, which enhances the robustness to occlusion and achieves remarkable results.

• We further improve the original dynamic routing algorithm by adaptively adjust the kernel size to alleviate computational burdens when capturing the landmark spatial positional relationship in the face image.

• We present a dual-attention mechanism to make the network automatically focus on the more advantageous features and suppress other unrelated ones.

• Experiment results show that the proposed DSCN achieves excellent performance on challenging benchmark datasets such as 300W, COFW and WFLW.

摘要

•We propose an hourglass capsule network to build a more accurate facial inter-feature spatial constrained model, which enhances the robustness to occlusion and achieves remarkable results.•We further improve the original dynamic routing algorithm by adaptively adjust the kernel size to alleviate computational burdens when capturing the landmark spatial positional relationship in the face image.•We present a dual-attention mechanism to make the network automatically focus on the more advantageous features and suppress other unrelated ones.•Experiment results show that the proposed DSCN achieves excellent performance on challenging benchmark datasets such as 300W, COFW and WFLW.

论文关键词：Face alignment,Hourglass capsule network,Adaptively local constrained dynamic routing,Capsule attention,Spatial attention

论文评审过程：Received 10 September 2019, Revised 29 August 2021, Accepted 31 August 2021, Available online 3 September 2021, Version of Record 20 September 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108297