arrow
返回

Gradient-Semantic Compensation for Incremental Semantic Segmentation

delete2024-01-01
delete6
delete
OA
AI
W
Wei Wei Cong
丛
丛杨 (Yang Cong) *
J
Jiahua Dong
G
Gan Sun
H
Henghui Ding
DOI:10.1109/TMM.2023.3336243delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Incremental semantic segmentation focuses on continually learning the segmentation of new coming classes without obtaining the training data from previously seen classes. However, most current methods fail to tackle catastrophic forgetting and background shift since they 1) treat all previous classes equally without considering different forgetting paces caused by imbalanced gradient back-propagation; 2) lack strong semantic guidance between classes. In this paper, to solve the aforementioned challenges, we propose a Gradient-Semantic Compensation (GSC) model, which surmounts incremental semantic segmentation from both gradient and semantic perspectives. Specifically, to handle catastrophic forgetting from the gradient aspect, we develop a step-aware gradient compensation that can balance forgetting paces of previously seen classes by re-weighting gradient back-propagation. Meanwhile, we propose a soft-sharp semantic relation distillation to distill consistent inter-class semantic relations via soft labels for alleviating catastrophic forgetting from the semantic aspect. In addition, we design a prototypical pseudo re-labeling which provides strong semantic guidance to mitigate background shift. It produces high-quality pseudo labels for background pixels belonging to previous classes by assessing distances of pixels relative to class-wise prototypes. Experiments on three public segmentation datasets provide strong evidence for the effectiveness of our proposed GSC model.
Keyword:
Gradient compensation
incremental learning
relations distillation
semantic segmentation

期刊

IEEE Transactions on Multimedia 封面图
IEEE Transactions on Multimedia
IF:
9.7
论文数:
4.6K
被引数:
2.4W

机构

U
university of chinese academy of sciences, cas
学者数:
4.1W
论文数: 3.8W
被引数: 75
C
chinese academy of sciences
学者数:
56.7W
论文数: 45.0W
被引数: 704
引用论文

引用论文

err分享
err收藏
err分享
err收藏
Learning Cross-Channel Representations for Semantic Segmentation
err2023-01-01
err28
PREAI
errMa, Lingfeng; Xie, Hongtao; Liu, Chuanbin; Zhang, Yongdong
err分享
err收藏
err分享
err收藏
Electrochemically assisted micro localized grafting of aptamers in a microchannel engraved in fluorinated thermoplastic polymer Dyneon THV
err2015-01-01
err0
PREAI
errC. Perréard; Y. Ladner; F. d'Orlyé; S. Descroix; V. Taniga; A. Varenne; F. Kanoufi; C. Slim; S. Griveau; F. Bedioui
err分享
err收藏
学者 查看更多内容