单幅图像人手参数模型深度网络回归方法  

Deep Neural Network Regression of MANO from a Single Image

作　　者：张惠嵌 李桂清[1] 柳雨新 聂勇伟 冼楚华[1] Zhang Huiqian;Li Guiqing;Liu Yuxin;Nie Yongwei;Xian Chuhua(School of Computer Science and Engineering,South China University of China,Guangzhou 510006)

机构地区：[1]华南理工大学计算机科学与工程学院,广州510006

出　　处：《计算机辅助设计与图形学学报》2022年第11期1684-1692,共9页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61972160,62072191);广东省自然科学基金(2021A1515012301,2019A1515010860);中央高校基本科研业务费(2020ZYGXZR042)。

摘　　要：为了提高从单目RGB图像估计人手的姿态与形状的精确性和准确性,加速3D重建过程,同时结合深度网络的高效性和迭代拟合的稳定性,提出单目图像人手网格重建方法.首先利用卷积神经网络模型从图像提取稀疏特征;然后根据稀疏特征回归人手模型参数,回归参数用于初始化迭代优化例程,将人手模型拟合到3D关节点上;最后用迭代拟合的人手参数逆向监督整个网络.基于弱透视投影模型,采用大型手势动作数据库FreiHand和ObMan进行实验的结果表明,所提方法在姿态误差和网格误差上较对比方法分别降低约52%和59%,在运行效率上较对比方法快约3倍.In order to improve the precision and efficiency of estimating the pose and shape of the 3D human hand from a monocular RGB image and accelerate 3D reconstruction,combining the efficiency of deep neural networks and accuracy of traditional optimization,DeepMANO is proposed.First,a DNN model is used to extract sparse features from image,then using these features to regress MANO model parameters.The regressed parameters are then used to initialize the iterative optimization routine which fits MANO to 3D joints.Finally,the parameters obtained by the routine are employed to supervise the network.Two large hand motion datasets,FreiHand and ObMan,are utilized to train the network.The approach assumes a weak perspective model.Experimental results show that the hand pose error is reduced by 52%and mesh error is reduced by 59%compared with state-of-the-art(SOTA)methods.The running time is about three times faster than SOTA methods.

关 键 词：人手姿态和形状 迭代优化 3D重建 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]