Improve image segmentation by identifying and relabeling out-of-candidate pixels with OCR
Improve image segmentation by identifying and relabeling out-of-candidate pixels with OCR
Out-of-Candidate Rectification for Weakly Supervised Semantic Segmentation
arXiv paper abstract https://arxiv.org/abs/2211.12268
arXiv PDF paper https://arxiv.org/pdf/2211.12268.pdf
Weakly supervised semantic segmentation is typically inspired by class activation maps, which serve as pseudo masks with class-discriminative regions highlighted.
… existing methods still commonly suffer from the unsolicited Out-of-Candidate (OC) error predictions that not belongs to the label candidates, which could be avoidable since the contradiction with image-level class tags is easy to be detected.
… develop a group ranking-based Out-of-Candidate Rectification (OCR) mechanism in a plug-and-play fashion.
Firstly, … adaptively split the semantic categories into In-Candidate (IC) and OC groups for each OC pixel according to their prior annotation correlation and posterior prediction correlation.
Then, … derive a differentiable rectification loss to force OC pixels to shift to the IC group.
… OCR … can achieve remarkable performance gains on both Pascal VOC … and MS COCO … datasets with negligible extra training overhead …
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
