Weakly supervised segmentation with maximum bipartite graph matching

• Main Authors: Liu, Weide, Zhang, Chi, Lin, Guosheng, Hung, Tzu-Yi, Miao, Chunyan
• 其他作者: School of Computer Science and Engineering
• 格式: Article Conference or Workshop Item
• 語言: English
• 出版: 2021
• 主題: Engineering::Computer science and engineering; Computer Vision; Graph Matching
• 在線閱讀: https://hdl.handle.net/10356/151821
• 機構: Nanyang Technological University

In the weakly supervised segmentation task with only image-level labels, a common step in many existing algorithms is first to locate the image regions corresponding to each existing class with the Class Activation Maps (CAMs), and then generate the pseudo ground truth masks based on the CAMs to tra...