发表机构
Center for Machine Vision and Signal Analysis (CMVS), University of Oulu; ELLIS Institute Finland; Zhejiang University（奥卢大学机器视觉与信号分析中心; 芬兰ELLIS研究所; 浙江大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对深度远程光电容积脉搏波描记法在头部姿态下性能下降问题，提出CanonicalPhys方法，通过添加可微四点单应性变换固定面部锚点到规范位置，利用规范框架表示相关先验，有效降低姿态影响，减少MAE退化。
AI 中文摘要
深度远程光电容积脉搏波描记法（rPPG）在正面、静止面部的心率误差可达到次bpm，但在头部姿态下性能会急剧下降。在MMPD数据集上，最先进的FactorizePhys主干网络的平均绝对误差（MAE）从正面（|偏航角|<15°）到大偏航角（|偏航角|≥45°）帧增长了1.60倍。我们认为姿态是一种“坐标结构”干扰，而非数据增强问题。我们引入了CanonicalPhys，它在前面添加了一个可微的四点单应性变换，将四个面部锚点固定在规范位置。在这个规范框架中，三个先验可以表示为每个像素的朗伯权重、跨感兴趣区域的时间一致性损失以及来自窗口化POS的知识蒸馏。在相同参数数量下，CanonicalPhys将MMPD数据集上从正面到大幅偏航的MAE退化从1.60倍降低到1.33倍，并将轻度偏航区间从1.32倍降低到1.07倍，在姿态丰富的目标上跨数据集MAE减少高达32%。
英文摘要
Deep remote photoplethysmography (rPPG) attains sub-bpm heart-rate error on frontal, stationary faces yet degrades sharply under head pose: on MMPD, the state-of-the-art FactorizePhys backbone's MAE grows $1.60\times$ from frontal ($|\text{yaw}|{<}15^\circ$) to large-yaw ($|\text{yaw}|{\geq}45^\circ$) frames. We argue that pose is a \emph{coordinate-structural} nuisance rather than a data-augmentation problem: in image coordinates the same pixel maps to different anatomy at different poses, blocking three priors otherwise natural for rPPG, namely the dichromatic reflection model, pulse-phase invariance across skin regions, and the POS/CHROM chromaticity projection, each of which presumes a stable anatomy-to-pixel mapping. We introduce \textbf{CanonicalPhys}, which prepends a differentiable four-point homography that fixes four facial anchors at canonical positions; in this canonical frame the three priors become expressible as a per-pixel Lambertian weight, a cross-ROI temporal consistency loss, and knowledge distillation from windowed POS, none of which adds trainable parameters over the backbone. At an identical parameter count, CanonicalPhys reduces MMPD's frontal-to-large-yaw MAE degradation from $1.60\times$ to $1.33\times$ and flattens the mild-yaw bin from $1.32\times$ to $1.07\times$ (across CanonicalPhys variants), with matched cross-dataset MAE reductions of up to $32\%$ on pose-rich targets. Code: https://github.com/infraface/CanonicalPhys
CommentsAccepted by IJCB 2026. Code: https://github.com/infraface/CanonicalPhys
