Research map / Detection and segmentation / Open-set and few-shot detection
Learning Open-Vocabulary Semantic Segmentation Models From Natural Language Supervision
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 masked entity completion to image-caption contrastive training (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 40.5 -> 48.9 | Helped |
| Add masked entity completion to image-caption contrastive training (ablation) | OVSegmentor · PASCAL Context | mIoU: 15.1 -> 19.9 | Helped |
| Add cross-image mask consistency to image-caption contrastive training (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 40.5 -> 44.8 | Helped |
| Add cross-image mask consistency to image-caption contrastive training (ablation) | OVSegmentor · PASCAL Context | mIoU: 15.1 -> 17.0 | Helped |
| Use both masked entity completion and cross-image mask consistency (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 40.5 -> 53.8 | Helped |
| Mask all frequent entities rather than a single entity (ablation) | OVSegmentor entity completion · PASCAL VOC 2012 | mIoU: 47.0 -> 48.9 | Helped |
| Mask visual entities rather than all nouns (ablation) | OVSegmentor entity completion · PASCAL VOC 2012 | mIoU: 45.4 -> 48.9 | Helped |
| Replace masked entity completion with MLM using group tokens (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 48.9 -> 42.8 | Hurt |
| Replace cross-image mask consistency with group consistency using NCE (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 53.8 -> 47.9 | Hurt |
| Increase group count from 8 to 16 at selection ratio 0.5 (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 53.6 -> 43.2 | Hurt |
| Initialize the encoders from DINO and BERT rather than CLIP (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 38.4 -> 40.5 | Helped |
| Use an EMA momentum model to generate pseudo-mask targets (ablation) | OVSegmentor · PASCAL VOC 2012 | mIoU: 50.9 -> 53.8 | Helped |
Closest papers by meaning
- Training-free Open-Vocabulary Semantic Segmentation via Diverse Prototype Construction and Sub-region Matching (AAAI 2025 · similarity 0.93)
- Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Models (CVPR 2024 · similarity 0.93)
- Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.91)
- Training-Free Open-Vocabulary Segmentation with Offline Diffusion-Augmented Prototype Generation (CVPR 2024 · similarity 0.91)
- Open-Vocabulary Segmentation with Semantic-Assisted Calibration (CVPR 2024 · similarity 0.91)
- Exploring Simple Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.91)
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.