Research map / Detection and segmentation / Open-set and few-shot detection
Stepping Out of Similar Semantic Space for Open-Vocabulary Segmentation
Yong Liu, Song-Li Wu, Sule Bai, Jiahao Wang, Yitong Wang, Yansong Tang
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 Gradient-Free Aggregation (ablation) | baseline OVS model · ADE-150 | mIoU: 33.1 -> 34.7 | Improved |
| add Gradient-Free Aggregation (ablation) | baseline OVS model · PC-459 | mIoU: 14.3 -> 16.0 | Improved |
| add Gradient-Free Aggregation (ablation) | baseline OVS model · OpenBench | mIoU: 42.3 -> 43.7 | Improved |
| add Proxy Calibration (ablation) | baseline OVS model · ADE-150 | mIoU: 33.1 -> 33.9 | Improved |
| add Proxy Calibration (ablation) | baseline OVS model · PC-459 | mIoU: 14.3 -> 17.2 | Improved |
| add Proxy Calibration (ablation) | baseline OVS model · OpenBench | mIoU: 42.3 -> 44.3 | Improved |
| add both GFA and PC (ablation) | baseline OVS model · PC-459 | mIoU: 14.3 -> 19.1 | Improved |
| add both GFA and PC (ablation) | baseline OVS model · OpenBench | mIoU: 42.3 -> 44.9 | Improved |
| use GFA instead of cross attention for fusion (ablation) | OVSNet · OpenBench | mIoU: 41.2 -> 43.7 | Improved |
| use GFA instead of self attention for fusion (ablation) | OVSNet · OpenBench | mIoU: 40.6 -> 43.7 | Improved |
| use GFA instead of cross attention for fusion (ablation) | OVSNet · PC-459 | mIoU: 14.8 -> 16.0 | Improved |
| use Beta gamma=2 rather than 0.5 in proxy calibration (ablation) | OVSNet PC · OpenBench | mIoU: 43.7 -> 44.3 | Improved |
Closest papers by meaning
- Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Models (CVPR 2024 · similarity 0.92)
- Transferable and Principled Efficiency for Open-Vocabulary Segmentation (CVPR 2024 · similarity 0.92)
- Training-free Open-Vocabulary Semantic Segmentation via Diverse Prototype Construction and Sub-region Matching (AAAI 2025 · similarity 0.92)
- Open-Vocabulary Segmentation with Semantic-Assisted Calibration (CVPR 2024 · similarity 0.91)
- OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts (NeurIPS 2025 · similarity 0.90)
- Renovating Names in Open-Vocabulary Segmentation Benchmarks (NeurIPS 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.
