A Simple Framework for Text-Supervised Semantic Segmentation

Muyang Yi, Quan Cui, Hao Wu, Cheng Yang, O. Yoshie +1 more
2/14/2026

No official code found. Best implementation: muyangyi/SimSeg (59 stars, MIT, last commit 2025-01-26).

Abstract

Text-supervised semantic segmentation is a novel research topic that allows semantic segments to emerge with image-text contrasting. However, pioneering methods could be subject to specifically designed network architectures. This paper shows that a vanilla contrastive language-image pretraining (CLIP) model is an effective text-supervised semantic segmentor by itself. First, we reveal that a vanilla CLIP is inferior to localization and segmentation due to its optimization being driven by densely aligning visual and language representations. Second, we propose the locality-driven alignment (LoDA) to address the problem, where CLIP optimization is driven by sparsely aligning local representations. Third, we propose a simple segmentation (SimSeg) framework. LoDA and SimSeg jointly amelio-rate a vanilla CLIP to produce impressive semantic segmentation results. Our method outperforms previous state-of-the-art methods on PASCAL VOC 2012, PASCAL Context and COCO datasets by large margins. Code and models are available at github.com/muyangyi/SimSeg.

DOISemantic Scholar

Code Implementations(1)

muyangyi/SimSegCommunity62%
MIT

[CVPR'23] A Simple Framework for Text-Supervised Semantic Segmentation

593PythonMar 22, 20231 years agoMIT

Similar Papers

Image-Text Co-Decomposition for Text-Supervised Semantic Segmentation

90% match1 repo222026

MixReorg: Cross-Modal Mixed Patch Reorganization is a Good Mask Learner for Open-World Semantic Segmentation

90% match2026

DIAL: Dense Image-text ALignment for Weakly Supervised Semantic Segmentation

88% match2026

UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation

87% match1 repo2122024

Structure-Aware Feature Rectification with Region Adjacency Graphs for Training-Free Open-Vocabulary Semantic Segmentation

87% match2025

SemiDAViL: Semi-supervised Domain Adaptation with Vision-Language Guidance for Semantic Segmentation

87% match1 repo102026

Cite this paper

@article{yi2026simple,
  title  = {A Simple Framework for Text-Supervised Semantic Segmentation},
  author = {Muyang Yi and Quan Cui and Hao Wu and Cheng Yang and O. Yoshie and Hongtao Lu},
  year   = {2026},
  doi    = {10.1109/CVPR52729.2023.00683},
  url    = {https://doi.org/10.1109/CVPR52729.2023.00683},
  journal = {CVPR 2023 2023}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/15e61063ddefcbe46d1ad5a1f4ce2e4a55e1a2ec.svg)](https://codeofpaper.com/paper/15e61063ddefcbe46d1ad5a1f4ce2e4a55e1a2ec)

Discussion