JMNet: A joint matting network for automatic human matting  

作　　者：Xian Wu Xiao-Nan Fang Tao Chen Fang-Lue Zhang 

机构地区：[1]Department of Computer Science and Technology,Tsinghua University,Beijing,100084,China [2]AI Center at Visual China Group,Burlingame,CA,94010,USA [3]School of Engineering and Computer Science,Victoria University of Wellington,Wellington,New Zealand

出　　处：《Computational Visual Media》2020年第2期215-224,共10页计算可视媒体（英文版）

基　　金：supported by National Natural Science Foundation of China(Grant Nos.61561146393 and61521002);supported by a Victoria Early-Career Research Excellence Award。

摘　　要：We propose a novel end-to-end deep learning framework, the Joint Matting Network(JMNet), to automatically generate alpha mattes for human images.We utilize the intrinsic structures of the human body as seen in images by introducing a pose estimation module,which can provide both global structural guidance and a local attention focus for the matting task. Our network model includes a pose network, a trimap network, a matting network, and a shared encoder to extract features for the above three networks. We also append a trimap refinement module and utilize gradient loss to provide a sharper alpha matte. Extensive experiments have shown that our method outperforms state-of-theart human matting techniques;the shared encoder leads to better performance and lower memory costs.Our model can process real images downloaded from the Internet for use in composition applications.

关 键 词：alpha matting human images deep learning pose estimation 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]