The field of remote physiological measurement is rapidly advancing, driven by the development of novel techniques and technologies for non-invasive and contactless monitoring of physiological signals. A key direction of research is the integration of multimodal sensing and machine learning to improve the accuracy and robustness of remote physiological measurement systems. Notable progress is being made in the development of datasets and platforms for remote physiological measurement, which is essential for training and evaluating machine learning models. Another area of focus is the development of methods for test-time adaptation, which enables models to adapt to new environments and conditions without requiring retraining.

Some noteworthy papers in this area include: CAST-Phys, which presents a novel dataset for multi-modal remote physiological emotion recognition. PhysioEdge, which introduces a custom hardware platform for synchronized multi-modal biomedical monitoring. Robust and Generalizable Heart Rate Estimation via Deep Learning for Remote Photoplethysmography in Complex Scenarios, which proposes an end-to-end network for remote photoplethysmography. Not Only Consistency: Enhance Test-Time Adaptation with Spatio-temporal Inconsistency for Remote Physiological Measurement, which proposes a novel test-time adaptation strategy for remote physiological measurement tasks.