基于MobileViT改进的红外人体姿态估计算法  

Improved infrared human pose estimation algorithm based on MobileViT

作　　者：张文扬 徐召飞 刘晴 王科俊[2] 岳广辉 王水根 尚在飞 ZHANG Wenyang;XU Zhaofei;LIU Qing;WANG Kejun;YUE Guanghui;WANG Shuigen;SHANG Zaifei(Graduate School of Harbin Engineering University(Yantai),Yantai Shandong 264000,China;College of Automation,Harbin Engineering University,Harbin Helongjiang 150001,China;Yantai Arrow Photoelectric Technology Co.,Ltd.,Yantai Shandong 264000,China;School of Biomedical Engineering,Health Science Center,Shenzhen University,Shenzhen Guangdong 518060,China;Military Representative Office of Luzhuang in Yantai,Yantai Shandong 264000,China)

机构地区：[1]哈尔滨工程大学研究生院,山东烟台264000 [2]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001 [3]烟台艾睿光电科技有限公司,山东烟台264000 [4]深圳大学医学部生物医学工程学院,广东深圳518060 [5]陆装驻烟台地区军事代表室,山东烟台264000

出　　处：《太赫兹科学与电子信息学报》2024年第7期781-791,共11页Journal of Terahertz Science and Electronic Information Technology

基　　金：国家自然科学基金资助项目(ZR202103080141)。

摘　　要：人体姿态估计主要依赖于视觉图像信息捕获关节点从而获得肢体和躯干的全局姿态信息。目前,基于可见光的深度学习方法具备较高的检测精确度,但隐私泄露的风险限制了其实际应用。同成本的红外探测器虽更能突出人体目标,但因成像分辨力较低,图像质量差,导致检测精确度下降。受视觉Transformer的启发,本文引入MobileViT-FPN提取人体关键点,利用MobileViT捕捉局部关节点特征和全局关节点特征关系,然后使用固定模式噪声(FPN)在多尺度上聚合这些表征信息,结合改进的OpenPose对关键点进行聚类,输出估计结果。在关键点级联阶段,注意力机制使模型自适应关注感兴趣区域,增强对遮挡部位的恢复。实验表明,该方法可以实时检测变化尺度和部分遮挡的红外人体目标,准确描绘人体姿态。Human pose estimation primarily relies on capturing joint points from visual image information to obtain global posture information of limbs and torso.Currently,depth learning methods based on visible light have high detection accuracy,but the risk of privacy leakage limits their practical application.Infrared detectors of the same cost can highlight human targets more effectively,but due to their lower imaging resolution and poor image quality,the detection accuracy is reduced.Inspired by visual Transformers,this paper introduces MobileViT-FPN to extract key human body points,using MobileViT to capture the relationship between local and global joint features,and then using Fixed Pattern Noise(FPN)to aggregate these representational information at multiple scales.Combined with an improved OpenPose for key point clustering,the estimated results are outputted.In the key point cascading phase,the attention mechanism allows the model to adaptively focus on the area of interest,enhancing the recovery of occluded parts.Experiments show that this method can real-time detect infrared human targets with varying scales and partial occlusions,accurately depicting human posture.

关 键 词：红外人体姿态估计 MobileViT主干网络 OpenPose网络 固定模式噪声 

分 类 号：TN919.8[电子电信—通信与信息系统]