Research map / Vision-language and multimodal / CLIP and zero-shot
Rewrite Caption Semantics: Bridging Semantic Gaps for Language-Supervised Semantic Segmentation
Yun Xing, Jian Kang, Aoran Xiao, Jiahao Nie, Ling Shao, Shijian Lu
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 concept curation (CoCu) | GroupViT · zero-shot segmentation average, C3 | mIoU: 8.2 -> 13.1 | Improved |
| add concept curation (CoCu) | GroupViT · zero-shot segmentation average, C12 | mIoU: 20.2 -> 22.7 | Improved |
| add concept curation (CoCu) | GroupViT · zero-shot segmentation average, C3,C12,Y14 with LC | mIoU: 23.1 -> 27.7 | Improved |
| add concept curation (CoCu) | GroupViT · Cityscapes, pretrain C3 | mIoU: 8.1 -> 8.2 | No clear effect |
| add concept curation (CoCu) | GroupViT · Pascal VOC, C3,C12,Y14 without LC vs 4096-batch GroupViT | mIoU: 52.4 -> 49.7 | Hurt |
| add concept curation (CoCu) | GroupViT · COCO, C3,C12,Y14 vs 4096-batch GroupViT | mIoU: 24.3 -> 22.0 | Hurt |
| add local consistency (LC) (ablation) | CoCu · zero-shot segmentation average | mIoU: 26.8 -> 27.7 | Improved |
| language-driven expansion with naive ranking and sampling (ablation) | GroupViT · zero-shot segmentation average, C3 | mIoU: 8.2 -> 9.9 | Improved |
| replace language-driven with vision-driven expansion (ablation) | Model #1 · zero-shot segmentation average, C3 | mIoU: 9.9 -> 10.3 | Improved |
| add text-to-vision-guided ranking (ablation) | Model #2 · zero-shot segmentation average, C3 | mIoU: 10.3 -> 12.4 | Improved |
| add cluster-guided sampling (ablation) | Model #3 · zero-shot segmentation average, C3 | mIoU: 12.4 -> 13.1 | Improved |
| add concept curation (CoCu) | GroupViT · ImageNet-1K zero-shot, C12 | Acc@1: 34.9 -> 38.4 | Improved |
Closest papers by meaning
- A Simple Baseline for Open-Vocabulary Semantic Segmentation with Pre-trained Vision-Language Model (ECCV 2022 · similarity 0.90)
- Training-Free Open-Vocabulary Segmentation with Offline Diffusion-Augmented Prototype Generation (CVPR 2024 · similarity 0.90)
- Dual Semantic Guidance for Open Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.89)
- A Simple Framework for Open-Vocabulary Zero-Shot Segmentation (ICLR 2025 · similarity 0.89)
- Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Models (CVPR 2024 · similarity 0.89)
- Open-Vocabulary Segmentation with Semantic-Assisted Calibration (CVPR 2024 · 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.
