Research map / Detection and segmentation / Open-set and few-shot detection
Text-guided Visual Prompt DINO for Generic Segmentation
Yuchen Guan, Chong Sun, Canmiao Fu, Zhipeng Huang, Chun Yuan, Chen Li
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 text alignment in backbone encoder (ablation) | Prompt-DINO baseline · ADE20K (out-domain) | PQ: 25.0 -> 32.7 | Improved |
| add early fusion of visual prompts (ablation) | Prompt-DINO · ADE20K (out-domain) | PQ: 32.7 -> 34.7 | Improved |
| add order-aligned query selection loss (ablation) | Prompt-DINO · ADE20K (out-domain) | PQ: 34.7 -> 35.9 | Improved |
| add training data from the RAP dual-path data engine (ablation) | Prompt-DINO (visual prompt) · LVIS | box AP: 47.9 -> 59.9 | Improved |
| add data engine data (ablation) | Prompt-DINO (visual prompt) · ADE20K (out-domain) | box AP: 26.8 -> 32.7 | Improved |
| add RAP labels to QwenVL labels for SA-1B annotation (ablation) | Prompt-DINO (text prompt) · ADE20K (out-domain) | box AP: 20.4 -> 23.0 | Improved |
| add RAP labels to QwenVL labels for SA-1B annotation (ablation) | Prompt-DINO (text prompt) · COCO (in-domain) | box AP: 56.7 -> 58.1 | Improved |
| add cross-verification filtering to data engine (ablation) | data engine · annotation quality | annotation CLIP similarity: 0.16 -> 0.23 | Improved |
| use Prompt-DINO instead of APE on Cityscapes | Prompt-DINO · Cityscapes (out-domain) | PQ and mIoU: PQ 33.3 -> 34.3; mIoU 44.2 -> 52.2 (OpenSeeD-L PQ 41.4 higher) | Mixed |
Closest papers by meaning
- Just a Few Glances: Open-Set Visual Perception with Image Prompt Paradigm (AAAI 2025 · similarity 0.89)
- IPO: Interpretable Prompt Optimization for Vision-Language Models (NeurIPS 2024 · similarity 0.89)
- Visual In-Context Prompting (CVPR 2024 · similarity 0.89)
- Learning to Prompt with Text Only Supervision for Vision-Language Models (AAAI 2025 · similarity 0.88)
- DPSeg: Dual-Prompt Cost Volume Learning for Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.88)
- Concept-Guided Prompt Learning for Generalization in Vision-Language Models (AAAI 2024 · similarity 0.88)
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.
