Key result
Dual-bridging networks with adversarial noise generation improve generalization and robustness of remote photoplethysmography models.
Why the study?
Deep rPPG models may not generalize well to target test domains with unseen noise and distortions.
A novel dual-bridging network with adversarial noise generation improves the generalization of deep learning models for remote photoplethysmography estimation from facial videos.
May enhance rPPG model robustness across domains; leaves open prospective clinical validation before use.
The remote photoplethysmography (rPPG) technique can estimate pulse-related metrics (e.g. heart rate and respiratory rate) from facial videos and has a high potential for health monitoring. The latest deep rPPG methods can model in-distribution noise due to head motion, video compression, etc., and estimate high-quality rPPG signals under similar scenarios. However, deep rPPG models may not generalize well to the target test domain with unseen noise and distortions. In this paper, to improve the generalization ability of rPPG models, we propose a dual-bridging network to reduce the domain discrepancy by aligning intermediate domains and synthesizing the target noise in the source domain for better noise reduction. To comprehensively explore the target domain noise, we propose a novel adversarial noise generation in which the noise generator indirectly competes with the noise reducer. To further improve the robustness of the noise reducer, we propose hard noise pattern mining to encourage the generator to learn hard noise patterns contained in the target domain features. We evaluated the proposed method on three public datasets with different types of interferences. Under different crossdomain scenarios, the comprehensive results show the effectiveness of our method.
No takes yet. Share an insight, caveat, or question.
Du et al. (2023) studied Remote photoplethysmography (rPPG) estimation. Dual-bridging network with adversarial noise generation was evaluated on Model performance on three public datasets under cross-domain scenarios. A proposed dual-bridging network with adversarial noise generation improved the generalization ability and robustness of remote photoplethysmography models across different cross-domain scenarios.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: