节点文献

Fast portrait automatic segmentation with coarse-to-fine CNNs

免费订阅

【作者】 Zhang XijinLi RuilongZhang Songhai

【Author】 Zhang Xijin;Li Ruilong;Zhang Songhai;Department of Computer Science and Technology, Tsinghua University;

【机构】 Department of Computer Science and Technology, Tsinghua University

【摘要】 In this paper, we propose a coarse-to-fine convolutional network framework designed with problem specific knowledge for fast automatic portrait segmentation. We built up a dataset of 7 100 portrait images which are frames from personal live show videos. The proposed network includes a coarse network which can learn global information and a fine network which utilizes local information to refine the coarse output. Additionally, an auxiliary contour loss is introduced to help training the coarse network. The proposed framework shows higher accuracy than the widely-used fully convolutional network. With light-weight post-processing, the predicted foreground mask can be used in real-time portrait video editing tasks such as background replacement.

【基金】 Supported by the Natural Science Foundation of China(61521002,61373069);Research Grant of Beijing Higher Institution Engineering Research Center
【所属期刊栏目】 Applized technology of CAD/CAT/CAM/CG (2017年02期)
  • 【DOI】10.19583/j.1003-4951.2017.02.0039
  • 【分类号】TP391.41
  • 【下载频次】41
节点文献中: 

本文链接的文献网络图示:

浏览历史:
下载历史: