Efficient and Robust Multidimensional Attention in Remote Physiological Sensing through Target Signal Constrained Factorization

TL;DR

This paper introduces a novel attention mechanism called TSFM and a dual-branch 3D-CNN architecture, MMRPhys, that together improve the robustness and real-time performance of remote physiological sensing across diverse conditions.

Contribution

The paper proposes TSFM, a new multidimensional attention module with physiological constraints, and MMRPhys, an efficient multitask model that enhances generalization in remote vital sign estimation.

Findings

MMRPhys with TSFM outperforms state-of-the-art methods in cross-dataset evaluations.

The approach maintains real-time inference latency suitable for practical use.

The method demonstrates robustness to domain shifts like lighting, pose, and physiological variations.

Abstract

Remote physiological sensing using camera-based technologies offers transformative potential for non-invasive vital sign monitoring across healthcare and human-computer interaction domains. Although deep learning approaches have advanced the extraction of physiological signals from video data, existing methods have not been sufficiently assessed for their robustness to domain shifts. These shifts in remote physiological sensing include variations in ambient conditions, camera specifications, head movements, facial poses, and physiological states which often impact real-world performance significantly. Cross-dataset evaluation provides an objective measure to assess generalization capabilities across these domain shifts. We introduce Target Signal Constrained Factorization module (TSFM), a novel multidimensional attention mechanism that explicitly incorporates physiological signal characteristics as factorization constraints, allowing more precise feature extraction. Building on this innovation, we present MMRPhys, an efficient dual-branch 3D-CNN architecture designed for simultaneous multitask estimation of photoplethysmography (rPPG) and respiratory (rRSP) signals from multimodal RGB and thermal video inputs. Through comprehensive cross-dataset evaluation on five benchmark datasets, we demonstrate that MMRPhys with TSFM significantly outperforms state-of-the-art methods in generalization across domain shifts for rPPG and rRSP estimation, while maintaining a minimal inference latency suitable for real-time applications. Our approach establishes new benchmarks for robust multitask and multimodal physiological sensing and offers a computationally efficient framework for practical deployment in unconstrained environments. The web browser-based application featuring on-device real-time inference of MMRPhys model is available at https://physiologicailab.github.io/mmrphys-live