用深度学习评估车载近红外心率监测，发现静止时误差7.5次/分
Internship Report: Benchmark of Deep Learning-based Imaging PPG in Automotive Domain
- 基于近红外摄像头与深度学习模型实现非接触式心率监测
- 头部静止时平均误差7.5次/分，轻微晃动时升至16.6次/分
- 适合关注智能座舱健康监测的研究者或汽车AI开发者
基于影像的光电容积脉搏波描记法（iPPG）可用于驾驶过程中心率监测，有望通过持续评估驾驶员身体状况减少交通事故。近期，利用近红外（NIR）摄像头的深度学习iPPG方法受到关注。为理解其在车载场景中的挑战，我们基于MR-NIRP Car数据集，对一种基于NIR的深度学习方法进行了基准测试。实验结果表明，在驾驶员头部保持静止时，平均绝对误差（MAE）为7.5 bpm；在存在小幅运动时，该值上升至16.6 bpm。结果表明，尽管该方法具有潜力，但仍需改进以满足真实驾驶环境下的可靠性要求。
原文摘要 · Abstract (English)
Imaging photoplethysmography (iPPG) can be used for heart rate monitoring during driving, which is expected to reduce traffic accidents by continuously assessing drivers' physical condition. Deep learning-based iPPG methods using near-infrared (NIR) cameras have recently gained attention as a promising approach. To help understand the challenges in applying iPPG in automotive, we provide a benchmark of a NIR-based method using a deep learning model by evaluating its performance on MR-NIRP Car dataset. Experiment results show that the average mean absolute error (MAE) is 7.5 bpm and 16.6 bpm under drivers' heads keeping still or having small motion, respectively. These findings suggest that while the method shows promise, further improvements are needed to make it reliable for real-world driving conditions.
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。