Temporal Physiological Signals

Utilizing physiological patterns for real human signal identification

Our research in temporal physiological signals focuses on developing advanced remote photoplethysmography (rPPG) techniques for non-contact physiological measurement. These methods enable real-time monitoring of vital signs like heart rate and respiratory rate through conventional cameras, providing crucial physiological cues for emotion authenticity verification.

Research Focus: Developing robust, generalizable remote physiological measurement systems that can operate reliably under dynamic domain shifts and varying environmental conditions.

Key Contributions

  • Period-LLM Framework: Extending periodic capability of multimodal large language models for physiological signal analysis
  • Stable Continual Adaptation: Developing test-time adaptation frameworks for remote physiological measurement in dynamic domains
  • Self-supervised Pre-training: rPPG-MAE with masked autoencoders for improved remote physiological measurements
  • Latent Codebook Querying: CodePhys framework for robust video-based remote physiological measurement
  • Generalizable Measurement: Integration of explicit and implicit prior knowledge for advancing remote physiological measurement
  • Hemoglobin-Assistant Networks: HemNet for enhanced video-based remote photoplethysmography measurement
Period-LLM: Extending the Periodic Capability of Multimodal Large Language Model
Y. Zhang, H. Lu, Q. Hu, Y. Wang, K. Yuan, X. Liu, and K. Wu
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025
source code
To Remember, To Adapt, To Preempt: A Stable Continual Test-Time Adaptation Framework for Remote Physiological Measurement in Dynamic Domain Shifts
S. Chu, J. Shi, X. Cheng, H. Chen, X. Liu, G. Zhao
ACM International Conference on Multimedia (ACMMM), 2025
source code
rPPG-MAE: Self-supervised Pretraining with Masked Autoencoders for Remote Physiological Measurements
X. Liu, Y. Zhang, Z. Yu, H. Lu, H. Yue, and J. Yang
IEEE Transactions on Multimedia, 2024
source code
CodePhys: Robust Video-based Remote Physiological Measurement through Latent Codebook Querying
S. Chu, M. Xia, M. Yuan, X. Liu, T. Seppanen, G. Zhao, and J. Shi
IEEE Journal of Biomedical and Health Informatics, 2025
Advancing Generalizable Remote Physiological Measurement through the Integration of Explicit and Implicit Prior Knowledge
Y. Zhang, H. Lu, X. Liu*, Y. Chen, and K. Wu
IEEE Transactions on Image Processing, 2025
source code
HemNet: Hemoglobin-Assistant Network for Video-based Remote Photoplethysmography Measurement
R. Wu, W. Zhuo, J. Shi, X. Liu, L. Shen, Y. Gong, and G. Zhao
IEEE Transactions on Circuits and Systems for Video Technology, 2025
« Back to Research Framework