Research map / Detection and segmentation / Open-set and few-shot detection
Prompt-Free Universal Region Proposal Network
Qihong Tang, Changhan Liu, Shaofeng Zhang, Wenbin Li, Qi Fan, Yang Gao
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 SIA (ablation) | Grounding DINO · CD-FSOD | AR100: 52.9 -> 57.8 | Improved |
| add CSP on top of SIA (ablation) | PF-RPN · CD-FSOD | AR100: 57.8 -> 60.2 | Improved |
| add CG-QS on top of SIA (ablation) | PF-RPN · CD-FSOD | AR100: 57.8 -> 59.6 | Improved |
| add all three modules (ablation) | Grounding DINO · CD-FSOD | AR100: 52.9 -> 60.7 | Improved |
| add MoE router in SIA (ablation) | PF-RPN · CD-FSOD | AR100: 58.6 -> 60.7 | Improved |
| add MoE router in SIA (ablation) | PF-RPN · ODinW13 | AR100: 68.7 -> 76.5 | Improved |
| increase COCO training data from 1% to 5% (ablation) | PF-RPN · CD-FSOD | AR100: 57.2 -> 58.8 | Improved |
| increase COCO training data from 5% to 10% (ablation) | PF-RPN (with 5% ImageNet) · CD-FSOD | AR100: 60.7 -> 60.9 | No clear effect |
| add 5% ImageNet classification data (ablation) | PF-RPN (5% COCO) · CD-FSOD | AR100: 58.8 -> 60.7 | Improved |
| replace RPN with PF-RPN | DE-ViT · COCO | AP: 33.0 -> 36.7 | Improved |
| replace RPN with PF-RPN | CD-ViTO · CD-FSOD | AP: 29.6 -> 35.1 | Improved |
| replace class-name prompts with PF-RPN (prompt-free) | Grounding DINO ResNet-50 · CD-FSOD | AR100: 40.4 -> 45.6 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Cascade RPN — Cascade RPN: Delving into High-Quality Region Proposal Network with Adaptive Convolution (NeurIPS 2019) · better on 2 of 2 shared results
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
- CP-DETR: Concept Prompt Guide DETR Toward Stronger Universal Object Detection (AAAI 2025 · similarity 0.86)
- Recognize Any Regions (NeurIPS 2024 · similarity 0.85)
- Relax Image-Specific Prompt Requirement in SAM: A Single Generic Prompt for Segmenting Camouflaged Objects (AAAI 2024 · similarity 0.85)
- PromptDet: A Lightweight 3D Object Detection Framework with LiDAR Prompts (AAAI 2025 · similarity 0.85)
- Towards Single-Source Domain Generalized Object Detection via Causal Visual Prompts (NeurIPS 2025 · similarity 0.85)
- Just a Few Glances: Open-Set Visual Perception with Image Prompt Paradigm (AAAI 2025 · similarity 0.85)
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.
