Research on Key Point Detection of Defogging Human Body in Bathing Scene

在机器人辅助老年人淋浴的环境中，水雾的存在会掩盖关键身体点的视线，对精确人体关键点检测构成挑战。本研究提出了一种端到端模型，将人类关键点检测与去雾技术相结合。该模型旨在监测老年人在淋浴过程中的关键点，从而防止事故发生。为了评估我们方法的有效性，我们制作了一个具有细水雾的模拟数据集，并比较了各种模型。实验结果表明，与我们精选数据集上的其他模型相比，我们提出的模型实现了82.63%的平均精度均值(mAP)的显著提高，比基线提高了23%。此外，它还可以有效监测老年人的淋浴过程。这项研究具有实际意义，并有可能提高老年人的生活质量。In the context of robot-assisted elderly showering, the presence of water mist can obscure the view of critical body points, posing a challenge for accurate human keypoint detection. This study proposes an end-to-end model that integrates human critical point detection with dehazing techniques. The model aims to monitor critical points in elderly individuals during the showering process, thereby preventing accidents. To assess the effectiveness of our approach, we curated a simulated dataset with featuring fine water mist and compared various models. Experimental results demonstrate that our proposed model achieves a significantly improved mean Average Precision (mAP) of 82.63% compared to other models on our curated dataset, marking increased by 23%. Additionally, it effectively monitors the showering process of the elderly. This research holds practical implications and has the potential to enhance the quality of life for older adults.