计算机科学与探索2024，Vol.18Issue(9)：2384-2394,11.DOI:10.3778/j.issn.1673-9418.2307016
空间注意力与位置优化的三维人体姿态估计域适应算法
Domain Adaptation Algorithm for 3D Human Pose Estimation with Spatial Attention and Position Optimization
摘要
Abstract
Existing 3D human pose estimators perform well on a single dataset but are limited by the single pose structure of the training data,resulting in insufficient generalization to cross-domain experiments.Existing methods mitigate this deficiency by increasing pose diversity,but their generated poses often lack validity.Moreover,there is still a significant gap between the global positions of poses in the target and source domains.To address these issues,a spatial attention and global position optimization domain adaptation algorithm for 3D human pose estimation based on generative adversarial network(GAN)is proposed.The algorithm introduces a spatial node attention module to constrain the generator to produce more natural human poses,and combines it with a pose position correction mod-ule to drive the generated poses to align to the target data domain,thus solving the above domain adaptation prob-lem.In addition,in order to improve the stability of estimator training,an end-to-end stochastic hybrid training strat-egy is proposed so that the pose estimator can take into account the learning of new and old data information.As a generative domain adaptation method,this algorithm can be efficiently applied to various two-stage 3D human pose esti-mators.Through cross-scene experiments and cross-dataset experiments,the results show that the proposed algo-rithm achieves the current best performance on several benchmark datasets.Among them,in the 3DHP dataset,the MPJPE and AUC metrics of the proposed method are optimized by 1.7%and 1.4%compared with the optimal work,which verifies that the proposed algorithm can effectively improve the generalization of 3D human pose esti-mators.关键词
三维人体姿态估计/无监督域适应/生成对抗网络(GAN)/注意力机制Key words
3D human pose estimation/unsupervised domain adaptation/generative adversarial network(GAN)/attention mechanism分类
信息技术与安全科学引用本文复制引用
姜友鹏,华阳,宋晓宁..空间注意力与位置优化的三维人体姿态估计域适应算法[J].计算机科学与探索,2024,18(9):2384-2394,11.基金项目
国家社会科学基金(21&ZD166) （21&ZD166）
江苏省自然科学基金(BK20221535). This work was supported by the National Social Science Foundation of China(21&ZD166),and the Natural Science Foundation of Jiangsu Province(BK20221535). （BK20221535）
