English

Region-level Contrastive and Consistency Learning for Semi-Supervised Semantic Segmentation

Computer Vision and Pattern Recognition 2022-04-29 v1

Abstract

Current semi-supervised semantic segmentation methods mainly focus on designing pixel-level consistency and contrastive regularization. However, pixel-level regularization is sensitive to noise from pixels with incorrect predictions, and pixel-level contrastive regularization has memory and computational cost with O(pixel_num^2). To address the issues, we propose a novel region-level contrastive and consistency learning framework (RC^2L) for semi-supervised semantic segmentation. Specifically, we first propose a Region Mask Contrastive (RMC) loss and a Region Feature Contrastive (RFC) loss to accomplish region-level contrastive property. Furthermore, Region Class Consistency (RCC) loss and Semantic Mask Consistency (SMC) loss are proposed for achieving region-level consistency. Based on the proposed region-level contrastive and consistency regularization, we develop a region-level contrastive and consistency learning framework (RC^2L) for semi-supervised semantic segmentation, and evaluate our RC2^2L on two challenging benchmarks (PASCAL VOC 2012 and Cityscapes), outperforming the state-of-the-art.

Keywords

Cite

@article{arxiv.2204.13314,
  title  = {Region-level Contrastive and Consistency Learning for Semi-Supervised Semantic Segmentation},
  author = {Jianrong Zhang and Tianyi Wu and Chuanghao Ding and Hongwei Zhao and Guodong Guo},
  journal= {arXiv preprint arXiv:2204.13314},
  year   = {2022}
}

Comments

Accepted by IJCAI 2022 (Long Oral)

R2 v1 2026-06-24T11:01:07.608Z