|
Weakly Supervised Semantic Segmentation via Self-Supervised Destruction Learning
Jinlong Li, Zequn Jie, Xu Wang, Yu Zhou, Lin Ma, Jianming Jiang
Neurocomputing (NEUCOM), 2023.
Abstract: In this paper, we propose a novel “destruction learning” method via self-supervised manner,
producing the CAM attention maps better covering the whole object rather than only the most discriminative regions
as previous approaches. Region destruction mechanism is proposed to deliberately “destruct” the global structure in
both mid-level and low-level feature learning following jigsaw puzzle operation, for better local feature extraction of
the classification network.
|
|
Expansion and Shrinkage of Localization for Weakly-Supervised Semantic Segmentation
Jinlong Li, Zequn Jie, Xu Wang, Xiaolin Wei, Lin Ma
Neural Information Processing Systems (NeurIPS), Spotlight (1.7%) 2022.
[ArXiv Link]
[Code]
Abstract: We propose a new training pipeline to alleviate the partial localization
issue of the CAM in Weakly-supervised image semantic segmentation, ESOL, in a Divide-and-Conquer manner.
|
|
Weakly Supervised Semantic Segmentation via Progressive Patch Learning
Jinlong Li, Zequn Jie, Xu Wang, Yu Zhou, Xiaolin Wei, Lin Ma
IEEE Transactions on Multimedia (TMM), 2022.
[ArXiv Link]
[IEEE Trans Link]
[Code]
Abstract: We propose a new training pipeline to alleviate the partial localization issue of
the CAM in Weakly-supervised image semantic segmentation, PPL, in an iterative training manner.
|