Research map / Detection and segmentation / Open-set and few-shot detection
MRD: Multi-resolution Retrieval-Detection Fusion for High-Resolution Image Understanding
Fan Yang, Xingping Dong, Xin Yu, Wenhan Luo, Wei Liu, Kaihao Zhang
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 MRD to RAP (LLaVA-v1.5-7B) | RAP · V* Bench overall | accuracy (%): 91.1 -> 95.6 | Improved |
| add MRD to RAP (LLaVA-ov-0.5B) | RAP · HR-Bench 4K overall | accuracy (%): 61.3 -> 64.6 | Improved |
| add MRD to RAP (LLaVA-ov-0.5B) | RAP · HR-Bench 8K overall | accuracy (%): 63.5 -> 64.5 | Improved |
| add multi-resolution semantic fusion to RAP (ablation) | RAP · V* Bench (LLaVA-ov-0.5B) | overall accuracy (%): 83.6 -> 85.8 | Improved |
| add open-vocabulary detector fusion to RAP (ablation) | RAP · V* Bench (LLaVA-ov-0.5B) | overall accuracy (%): 83.6 -> 86.7 | Improved |
| add detector fusion to RAP (ablation) | RAP · V* Bench spatial (LLaVA-ov-0.5B) | accuracy (%): 84.2 -> 84.2 | No clear effect |
| add multi-resolution fusion on top of RAP+OVD (ablation) | RAP+OVD · V* Bench (LLaVA-ov-0.5B) | overall accuracy (%): 86.7 -> 88.9 | Improved |
| use standalone OVD instead of RAP retrieval (ablation) | OVD · V* Bench spatial (multi-object) | accuracy (%): 84.2 -> 82.9 | Hurt |
| add MRD fusion to cut search steps to 50 | RAP · HR-Bench 4K total time (v1.5-7B) | seconds: 63.4 -> 53.4 | Improved |
| add MRD | RAP · GPU memory (HR-Bench 4K, v1.5-7B) | GB: 21.2 -> 23.4 | Hurt |
Closest papers by meaning
- Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models (AAAI 2025 · similarity 0.88)
- Retrieval-Augmented Perception: High-resolution Image Perception Meets Visual RAG (ICML 2025 · similarity 0.88)
- CVSearch: Empowering Multimodal LLMs with Cognitive Visual Search for High-Resolution Image Perception (ICML 2026 · similarity 0.87)
- MMIU: Multimodal Multi-image Understanding for Evaluating Large Vision-Language Models (ICLR 2025 · similarity 0.87)
- RMIR: A Benchmark Dataset for Reasoning-Intensive Multimodal Image Retrieval (CVPR 2026 · similarity 0.86)
- Cyclic Contrastive Knowledge Transfer for Open-Vocabulary Object Detection (ICLR 2025 · similarity 0.86)
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.
