AMMSM: Adaptive Motion Magnification and Sparse Mamba for Micro-Expression Recognition
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916667873820672 |
|---|---|
| author | Liu, Xuxiong Dong, Tengteng Wang, Fei Feng, Weijie Sun, Xiao |
| author_facet | Liu, Xuxiong Dong, Tengteng Wang, Fei Feng, Weijie Sun, Xiao |
| contents | Micro-expressions are typically regarded as unconscious manifestations of a person's genuine emotions. However, their short duration and subtle signals pose significant challenges for downstream recognition. We propose a multi-task learning framework named the Adaptive Motion Magnification and Sparse Mamba (AMMSM) to address this. This framework aims to enhance the accurate capture of micro-expressions through self-supervised subtle motion magnification, while the sparse spatial selection Mamba architecture combines sparse activation with the advanced Visual Mamba model to model key motion regions and their valuable representations more effectively. Additionally, we employ evolutionary search to optimize the magnification factor and the sparsity ratios of spatial selection, followed by fine-tuning to improve performance further. Extensive experiments on two standard datasets demonstrate that the proposed AMMSM achieves state-of-the-art (SOTA) accuracy and robustness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_24057 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AMMSM: Adaptive Motion Magnification and Sparse Mamba for Micro-Expression Recognition Liu, Xuxiong Dong, Tengteng Wang, Fei Feng, Weijie Sun, Xiao Computer Vision and Pattern Recognition Micro-expressions are typically regarded as unconscious manifestations of a person's genuine emotions. However, their short duration and subtle signals pose significant challenges for downstream recognition. We propose a multi-task learning framework named the Adaptive Motion Magnification and Sparse Mamba (AMMSM) to address this. This framework aims to enhance the accurate capture of micro-expressions through self-supervised subtle motion magnification, while the sparse spatial selection Mamba architecture combines sparse activation with the advanced Visual Mamba model to model key motion regions and their valuable representations more effectively. Additionally, we employ evolutionary search to optimize the magnification factor and the sparsity ratios of spatial selection, followed by fine-tuning to improve performance further. Extensive experiments on two standard datasets demonstrate that the proposed AMMSM achieves state-of-the-art (SOTA) accuracy and robustness. |
| title | AMMSM: Adaptive Motion Magnification and Sparse Mamba for Micro-Expression Recognition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.24057 |