Self-similarity Prior Distillation for Unsupervised Remote Physiological Measurement
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917780733820928 |
|---|---|
| author | Zhang, Xinyu Sun, Weiyu Lu, Hao Chen, Ying Ge, Yun Huang, Xiaolin Yuan, Jie Chen, Yingcong |
| author_facet | Zhang, Xinyu Sun, Weiyu Lu, Hao Chen, Ying Ge, Yun Huang, Xiaolin Yuan, Jie Chen, Yingcong |
| contents | Remote photoplethysmography (rPPG) is a noninvasive technique that aims to capture subtle variations in facial pixels caused by changes in blood volume resulting from cardiac activities. Most existing unsupervised methods for rPPG tasks focus on the contrastive learning between samples while neglecting the inherent self-similar prior in physiological signals. In this paper, we propose a Self-Similarity Prior Distillation (SSPD) framework for unsupervised rPPG estimation, which capitalizes on the intrinsic self-similarity of cardiac activities. Specifically, we first introduce a physical-prior embedded augmentation technique to mitigate the effect of various types of noise. Then, we tailor a self-similarity-aware network to extract more reliable self-similar physiological features. Finally, we develop a hierarchical self-distillation paradigm to assist the network in disentangling self-similar physiological patterns from facial videos. Comprehensive experiments demonstrate that the unsupervised SSPD framework achieves comparable or even superior performance compared to the state-of-the-art supervised methods. Meanwhile, SSPD maintains the lowest inference time and computation cost among end-to-end models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2311_05100 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Self-similarity Prior Distillation for Unsupervised Remote Physiological Measurement Zhang, Xinyu Sun, Weiyu Lu, Hao Chen, Ying Ge, Yun Huang, Xiaolin Yuan, Jie Chen, Yingcong Computer Vision and Pattern Recognition Multimedia Remote photoplethysmography (rPPG) is a noninvasive technique that aims to capture subtle variations in facial pixels caused by changes in blood volume resulting from cardiac activities. Most existing unsupervised methods for rPPG tasks focus on the contrastive learning between samples while neglecting the inherent self-similar prior in physiological signals. In this paper, we propose a Self-Similarity Prior Distillation (SSPD) framework for unsupervised rPPG estimation, which capitalizes on the intrinsic self-similarity of cardiac activities. Specifically, we first introduce a physical-prior embedded augmentation technique to mitigate the effect of various types of noise. Then, we tailor a self-similarity-aware network to extract more reliable self-similar physiological features. Finally, we develop a hierarchical self-distillation paradigm to assist the network in disentangling self-similar physiological patterns from facial videos. Comprehensive experiments demonstrate that the unsupervised SSPD framework achieves comparable or even superior performance compared to the state-of-the-art supervised methods. Meanwhile, SSPD maintains the lowest inference time and computation cost among end-to-end models. |
| title | Self-similarity Prior Distillation for Unsupervised Remote Physiological Measurement |
| topic | Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2311.05100 |