Research map / Detection and segmentation / Open-set and few-shot detection
Visual Semantic Information Pursuit: A Survey
Daqi Liu, Miroslaw Z. Bober, Josef Kittler
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 visual context reasoning (SIN) to detector | object detection · VOC 2007 test | mAP: 73.2 -> 76.0 | Improved |
| add visual context reasoning (ION) to detector | object detection · COCO 2015 test-dev | AP: 21.1 -> 23.0 | Improved |
| add visual context reasoning (SIN) | object detection · COCO 2015 test-dev | AP: 21.1 -> 23.2 | Improved |
| add CRF to DAG-RNN | semantic segmentation · Pascal Context | mIOU: 42.6 -> 43.7 | Improved |
| add CRF to DAG-RNN | semantic segmentation · COCO Stuff | mIOU: 30.4 -> 31.2 | Improved |
| add CRF to DAG-RNN | semantic segmentation · Sift Flow | mIOU: 44.4 -> 44.8 | Improved |
| add CRF to DeepLab | semantic segmentation · Pascal Context | mIOU: 37.6 -> 39.6 | Improved |
| add DAG-RNN higher-order context instead of FCN-8s | semantic segmentation · COCO Stuff | mIOU: 27.2 -> 30.4 | Improved |
| add SCA-M to CAI | visual relationship detection · Visual Relationship dataset, relationship R@50 | R@50: 15.63 -> 19.54 | Improved |
| use weakly-supervised instead of fully-supervised relationship detection | visual relationship detection · Visual Relationship dataset, predicate R@50 | R@50: 47.43 -> 55.16 | Hurt |
| use Zoom-Net hierarchy reasoning instead of DR-Net | visual relationship detection · Visual Genome (k=100), relationship R@50 | R@50: 12.56 -> 19.97 | Improved |
| use Zoom-Net instead of ViP | visual relationship detection · Visual Genome (k=100), phrase R@50 | R@50: 15.70 -> 20.84 | Improved |
Closest papers by meaning
- VisRes Bench: On Evaluating the Visual Reasoning Capabilities of VLMs (CVPR 2026 · similarity 0.86)
- VLScene: Vision-Language Guidance Distillation for Camera-Based 3D Semantic Scene Completion (AAAI 2025 · similarity 0.86)
- VidSeg: Training-free Video Semantic Segmentation based on Diffusion Models (CVPR 2025 · similarity 0.85)
- Lexicon3D: Probing Visual Foundation Models for Complex 3D Scene Understanding (NeurIPS 2024 · similarity 0.85)
- CVSearch: Empowering Multimodal LLMs with Cognitive Visual Search for High-Resolution Image Perception (ICML 2026 · similarity 0.85)
- Mind the GAP: Glimpse-based Active Perception improves generalization and sample efficiency of visual reasoning (ICLR 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.
