首次用贝叶斯网络提升远端光体积描记的可靠性，精准估计心跳测量不确定性。
RF-BayesPhysNet: A Bayesian rPPG Uncertainty Estimation Method for Complex Scenarios
- 引入贝叶斯神经网络建模测量中的随机与认知不确定性。
- 在UBFC-RPPG数据集上MAE达2.56，仅多出两倍参数量。
- 开源代码，适合需要高可信度心跳检测的医疗与可穿戴场景。
远端光体积描记（rPPG）通过摄像头捕捉面部皮肤微弱颜色变化来推断心率，在非接触式心率监测中潜力巨大。然而在光照变化、头部移动等复杂场景下，测量精度显著下降，远低于理想实验室条件。现有深度学习模型普遍忽略测量不确定性的量化，限制了其在动态环境中的可信度。为此，本文首次将贝叶斯神经网络引入rPPG领域，提出鲁棒融合贝叶斯生理网络（RF-BayesPhysNet），可同时建模随机不确定性（aleatoric）与认知不确定性（epistemic）。采用变分推断平衡精度与计算效率。由于当前rPPG领域缺乏不确定性评估指标，本文提出使用斯皮尔曼相关系数、预测区间覆盖率及置信区间宽度，评估不同噪声条件下不确定性估计的有效性。实验表明，该模型仅比传统网络多出两倍参数，就在UBFC-RPPG数据集上达到2.56的平均绝对误差（MAE），在无噪与低噪条件下表现出优异的不确定性估计能力，提供预测置信度，显著增强真实应用中的鲁棒性。代码已开源：https://github.com/AIDC-rPPG/RF-Net。
原文摘要 · Abstract (English)
Remote photoplethysmography (rPPG) technology infers heart rate by capturing subtle color changes in facial skin using a camera, demonstrating great potential in non-contact heart rate measurement. However, measurement accuracy significantly decreases in complex scenarios such as lighting changes and head movements compared to ideal laboratory conditions. Existing deep learning models often neglect the quantification of measurement uncertainty, limiting their credibility in dynamic scenes. To address the issue of insufficient rPPG measurement reliability in complex scenarios, this paper introduces Bayesian neural networks to the rPPG field for the first time, proposing the Robust Fusion Bayesian Physiological Network (RF-BayesPhysNet), which can model both aleatoric and epistemic uncertainty. It leverages variational inference to balance accuracy and computational efficiency. Due to the current lack of uncertainty estimation metrics in the rPPG field, this paper also proposes a new set of methods, using Spearman correlation coefficient, prediction interval coverage, and confidence interval width, to measure the effectiveness of uncertainty estimation methods under different noise conditions. Experiments show that the model, with only double the parameters compared to traditional network models, achieves a MAE of 2.56 on the UBFC-RPPG dataset, surpassing most models. It demonstrates good uncertainty estimation capability in no-noise and low-noise conditions, providing prediction confidence and significantly enhancing robustness in real-world applications. We have open-sourced the code at https://github.com/AIDC-rPPG/RF-Net
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。