Research map / Detection and segmentation / Open-set and few-shot detection
S2C2Seg: Semantic-Spatial Consistency and Category Optimization for Open-Vocabulary Segmentation
Yuhao Qing, Yueying Wang, Chaoyang Chen, Weidong Zhang, Jie Wen, Xin Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add S2C2Seg to SCLIP | SCLIP · 8-benchmark average (training-free OVSS) | mIoU (avg): 38.2 -> 47.9 | Improved |
| add S2C2Seg to ProxyCLIP | ProxyCLIP · 8-benchmark average (training-free OVSS) | mIoU (avg): 42.3 -> 49.1 | Improved |
| add S2C2Seg to CorrCLIP | CorrCLIP · 8-benchmark average (training-free OVSS) | mIoU (avg): 47.8 -> 51.2 | Improved |
| add CSS category subset selection (ablation) | S2C2Seg (CSG only) · 8-benchmark average (CSS scoring ablation) | mIoU (avg): 45.9 -> 49.1 | Improved |
| add CSG on top of CSS (ablation) | S2C2Seg · VOC21 | mIoU: 70.4 -> 72.8 | Improved |
| add CSS on top of CSG (ablation) | S2C2Seg · ADE20K | mIoU: 21.7 -> 23.1 | Improved |
| add CSS (w/o CSG variant) on ADE20K (ablation) | ProxyCLIP · ADE20K | mIoU: 20.2 -> 20.4 | No clear effect |
| add spatial score to global score in CSS (ablation) | S2C2Seg · 8-benchmark average (CSS scoring ablation) | mIoU (avg): 45.9 -> 47.7 | Improved |
| use category-aware fusion instead of additive (ablation) | S2C2Seg · Context59 | mIoU: 44.1 -> 48.7 | Improved |
| use multiplicative instead of additive fusion (ablation) | S2C2Seg · Context59 | mIoU: 44.1 -> 42.8 | Hurt |
| vary lambda in 0.4-0.8, tau in 0.2-0.5, Kmax in 15-25 (ablation) | S2C2Seg · VOC (ProxyCLIP) | mIoU: 48.0-49.5 | No clear effect |
| use oracle ground-truth categories instead of CSS (ablation) | S2C2Seg · 8-benchmark average (CSS scoring ablation) | mIoU (avg): 49.1 -> 61.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- ProxyCLIP — ProxyCLIP: Proxy Attention Improves CLIP for Open-Vocabulary Segmentation (ECCV 2024) · better on 5 of 5 shared results
- CorrCLIP — CorrCLIP: Reconstructing Patch Correlations in CLIP for Open-Vocabulary Semantic Segmentation (ICCV 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Exploring Simple Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.91)
- Open-Vocabulary Segmentation with Semantic-Assisted Calibration (CVPR 2024 · similarity 0.89)
- Unveiling the Knowledge of CLIP for Training-Free Open-Vocabulary Semantic Segmentation (AAAI 2025 · similarity 0.89)
- OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts (NeurIPS 2025 · similarity 0.89)
- Effective SAM Combination for Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.89)
- Dual Semantic Guidance for Open Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
