基于非全局依赖积分回归的轻量姿态估计网络  

Lightweight pose estimation network based on non-globally dependent integral regression

作　　者：佘本杰 苏树智 朱彦敏 华健 王超 SHE Benjie;SU Shuzhi;ZHU Yanmin;HUA Jian;WANG Chao(School of Computer Science and Engineering,Anhui University of Science and Technology,Huainan Anhui 232001,China;School of Mechatronics Engineering,Anhui University of Science and Technology,Huainan Anhui 232001,China)

机构地区：[1]安徽理工大学计算机科学与工程学院,安徽淮南232001 [2]安徽理工大学机电工程学院,安徽淮南232001

出　　处：《计算机应用》2025年第3期972-977,共6页journal of Computer Applications

基　　金：国家自然科学基金资助项目(52374155);安徽省高等学校自然科学研究项目(重大项目)(2022AH040113);安徽省自然科学基金资助项目(2308085MF218);安徽理工大学研究生创新基金资助项目(2023cx2132)。

摘　　要：基于热图检测的人体姿态估计网络取得了巨大的成功,然而由于冗余计算、量化误差以及热图解码的需求,基于热图检测的方法参数量较大。针对上述问题,设计基于非全局依赖积分回归的轻量姿态估计网络(Lite-NIRNet)。Lite-NIRNet通过局部卷积(PConv)降低网络的冗余计算,从而使网络更加轻量。为缓解PConv导致的信息丢失问题,引入坐标注意力(CA)机制融合跨通道特征,以提升网络性能。此外,设计非全局依赖的积分回归(NIR)模块,通过该模块,网络可以融入坐标进行监督,从而减少量化误差对网络性能的影响。所提的NIR可有效降低传统积分回归在期望计算时产生的偏差,从而兼顾更好的学习梯度和更低的偏差。实验结果表明,Lite-NIRNet与较先进的高分辨率网络(HRNet)相比,在COCO验证集上的参数量和计算量分别降低了73.0%和63.4%,平均精度均值(mAP)不需要额外的热图解码即可达到72.8%;在MPII验证集上,Lite-NIRNet在网络性能和复杂度之间也能实现良好的平衡。Significant success has been achieved in human pose estimation networks based on heatmap detection.However,the methods based on heatmap detection has a large number of parameters due to redundant computations,quantization errors,and the requirement of heatmap decoding.To address these issues,a Lightweight pose estimation Network based on Non-globally dependent Integral Regression(Lite-NIRNet)was designed to reduce redundant computations in the network by employing Partial Convolution(PConv),which made the network more lightweight.To respond to the information loss caused by PConv,a Coordinate Attention(CA)mechanism was introduced to fuse interchannel features,thereby enhancing the network performance.Additionally,a Non-globally dependent Integral Regression(NIR)module was designed to incorporate coordinate supervision to the network,which reduced the influence of quantization errors on network performance.The proposed NIR was able to reduce the bias produced by traditional integral regression during expectation calculations effectively,balancing better learning gradients with lower bias.Experimental results show that compared with the advanced High-Resolution Network(HRNet),Lite-NIRNet reduces the number of parameters and computational complexity by 73.0%and 63.4%,respectively,on COCO validation set,and achieves the mean Average Precision(mAP)of 72.8%without additional heatmap decoding.Furthermore,on MPII validation set,Lite-NIRNet can also achieve a good balance between network performance and complexity.

关 键 词：姿态估计 高分辨率网络 局部卷积 坐标注意力 积分回归 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]