IEEE J Biomed Health Inform. 2024 Aug;28(8):4613-4624. doi: 10.1109/JBHI.2024.3400869. Epub 2024 Aug 6.
Remote photoplethysmography (rPPG) is a non-contact method that employs facial videos for measuring physiological parameters. Existing rPPG methods have achieved remarkable performance. However, the success mainly profits from supervised learning over massive labeled data. On the other hand, existing unsupervised rPPG methods fail to fully utilize spatio-temporal features and encounter challenges in low-light or noise environments. To address these problems, we propose an unsupervised contrast learning approach, ST-Phys. We incorporate a low-light enhancement module, a temporal dilated module, and a spatial enhanced module to better deal with long-term dependencies under the random low-light conditions. In addition, we design a circular margin loss, wherein rPPG signals originating from identical videos are attracted, while those from distinct videos are repelled. Our method is assessed on six openly accessible datasets, including RGB and NIR videos. Extensive experiments reveal the superior performance of our proposed ST-Phys over state-of-the-art unsupervised rPPG methods. Moreover, it offers advantages in parameter reduction and noise robustness.
远程光体积描记术（rPPG）是一种非接触式方法，利用面部视频来测量生理参数。现有的 rPPG 方法已经取得了显著的性能。然而，成功主要得益于对大量标记数据的监督学习。另一方面，现有的无监督 rPPG 方法未能充分利用时空特征，并且在低光或噪声环境中遇到挑战。为了解决这些问题，我们提出了一种无监督对比学习方法 ST-Phys。我们结合了低光增强模块、时间扩张模块和空间增强模块，以更好地处理随机低光条件下的长期依赖关系。此外，我们设计了一种圆形边界损失，其中来自相同视频的 rPPG 信号被吸引，而来自不同视频的信号则被排斥。我们的方法在六个公开可用的数据集上进行了评估，包括 RGB 和 NIR 视频。广泛的实验表明，我们提出的 ST-Phys 优于最先进的无监督 rPPG 方法。此外，它在参数减少和噪声鲁棒性方面具有优势。
