核心概念
SeCo proposes a 'Separate and Conquer' scheme to address the co-occurrence issue in weakly supervised semantic segmentation.
統計資料
"Extensive experiments are conducted, validating the efficiency of our method tackling co-occurrence and the superiority over previous single-staged and even multi-staged competitors on PASCAL VOC and MS COCO."
引述
"Essentially, the co-occurrence problem arises because the co-occurring contexts coupled in images confuse networks and incur semantic bias during the feature representation, resulting in false positive pixels activated with high probability."