Contrastive Prompt Clustering for Weakly Supervised Semantic Segmentation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914015002755072 |
|---|---|
| author | Wu, Wangyu Chen, Zhenhong Ma, Xiaowen Zhang, Wenqiao Qiu, Xianglin Song, Siqi Huang, Xiaowei Ma, Fei Xiao, Jimin |
| author_facet | Wu, Wangyu Chen, Zhenhong Ma, Xiaowen Zhang, Wenqiao Qiu, Xianglin Song, Siqi Huang, Xiaowei Ma, Fei Xiao, Jimin |
| contents | Weakly Supervised Semantic Segmentation (WSSS) with image-level labels has gained attention for its cost-effectiveness. Most existing methods emphasize inter-class separation, often neglecting the shared semantics among related categories and lacking fine-grained discrimination. To address this, we propose Contrastive Prompt Clustering (CPC), a novel WSSS framework. CPC exploits Large Language Models (LLMs) to derive category clusters that encode intrinsic inter-class relationships, and further introduces a class-aware patch-level contrastive loss to enforce intra-class consistency and inter-class separation. This hierarchical design leverages clusters as coarse-grained semantic priors while preserving fine-grained boundaries, thereby reducing confusion among visually similar categories. Experiments on PASCAL VOC 2012 and MS COCO 2014 demonstrate that CPC surpasses existing state-of-the-art methods in WSSS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_17009 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Contrastive Prompt Clustering for Weakly Supervised Semantic Segmentation Wu, Wangyu Chen, Zhenhong Ma, Xiaowen Zhang, Wenqiao Qiu, Xianglin Song, Siqi Huang, Xiaowei Ma, Fei Xiao, Jimin Computer Vision and Pattern Recognition Weakly Supervised Semantic Segmentation (WSSS) with image-level labels has gained attention for its cost-effectiveness. Most existing methods emphasize inter-class separation, often neglecting the shared semantics among related categories and lacking fine-grained discrimination. To address this, we propose Contrastive Prompt Clustering (CPC), a novel WSSS framework. CPC exploits Large Language Models (LLMs) to derive category clusters that encode intrinsic inter-class relationships, and further introduces a class-aware patch-level contrastive loss to enforce intra-class consistency and inter-class separation. This hierarchical design leverages clusters as coarse-grained semantic priors while preserving fine-grained boundaries, thereby reducing confusion among visually similar categories. Experiments on PASCAL VOC 2012 and MS COCO 2014 demonstrate that CPC surpasses existing state-of-the-art methods in WSSS. |
| title | Contrastive Prompt Clustering for Weakly Supervised Semantic Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.17009 |