← 研踪 PaperTrailarXiv:2611.26367 · v1
Contactless Estimation of Respiratory Frequency Using 3D-CNN on Thermal Images
首次提交 2026-11-20 00:33
版本更新 2026-11-20 00:33
雷达观测 2026-11-20 12:52
摘要
Monitoring physiological parameters such as respiratory rate () is essential for diagnosing and managing various pathological conditions. Thermal imaging offers a promising contactless alternative to traditional methods, which often rely on partially invasive sensors or obtrusive wearable systems. However, existing approaches for estimation from thermal signals typically require extensive pre-processing and manual or semi-automatic region-of-interest (ROI) tracking, limiting their practical applicability. This study proposes a deep learning-based method for estimating directly from thermal videos, eliminating the need for complex pre-processing and ROI tracking. A 3D Convolutional Neural Network (3D-CNN) is developed to operate on raw thermal video data. To address challenges related to small datasets, the model is trained using data augmentation and transfer learning from synthetic datasets. Experimental results demonstrate that the proposed approach achieves a validation score of approximately 0.61 on both pre-processed and raw thermal videos. By simplifying the workflow, this method holds promise for enhancing the feasibility of thermal imaging in real-world applications, such as remote healthcare and driver monitoring in automotive applications.
期刊信息:IEEE Journal of Biomedical and Health Informatics, 29(10), 7387-7395, 2025
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-20 00:33 · 最新观测版本
Contactless Estimation of Respiratory Frequency Using 3D-CNN on Thermal Images
Monitoring physiological parameters such as respiratory rate () is essential for diagnosing and managing various pathological conditions. Thermal imaging offers a promising contactless alternative to traditional methods, which often rely on partially invasive sensors or obtrusive wearable systems. However, existing approaches for estimation from thermal signals typically require extensive pre-processing and manual or semi-automatic region-of-interest (ROI) tracking, limiting their practical applicability. This study proposes a deep learning-based method for estimating directly from thermal videos, eliminating the need for complex pre-processing and ROI tracking. A 3D Convolutional Neural Network (3D-CNN) is developed to operate on raw thermal video data. To address challenges related to small datasets, the model is trained using data augmentation and transfer learning from synthetic datasets. Experimental results demonstrate that the proposed approach achieves a validation score of approximately 0.61 on both pre-processed and raw thermal videos. By simplifying the workflow, this method holds promise for enhancing the feasibility of thermal imaging in real-world applications, such as remote healthcare and driver monitoring in automotive applications.
查看此版本解读 →查看此版本原文 ↗