Research map / Detection and segmentation / Open-set and few-shot detection
Beyond Holistic Object Recognition: Enriching Image Understanding With Part States
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 |
|---|---|---|---|
| Combine global RGB context with local part regions and learn them jointly | ISIN · PASCAL VOC 2010 Part State Dataset | mAP: 32.3 -> 46.5 | Helped |
| Add iterative state-guided S-image updates instead of using one iteration (ablation) | ISIN · PASCAL VOC 2010 Part State Dataset | mAP: 40.7 -> 46.5; 5% | Helped |
| Train the unfolded architecture rather than the iterative framework (ablation) | ISIN · PASCAL VOC 2010 Part State Dataset | mAP: 46.5 -> 46.9; 0.4% | Helped |
| Use ground-truth part segmentation (ablation) | ISIN · PASCAL VOC 2010 Part State Dataset | mAP: 46.5 -> 61.5 | Helped |
| Add more iterative refinement steps (ablation) | ISIN · PASCAL VOC 2010 Part State Dataset | mAP: 40.7 -> 46.5 | Helped |
| Use part-state features for relationship prediction | Visual relationship detection · Visual relationship dataset | Relationship detection R@100: 9.04 -> 26.13 | Helped |
| Use part-state features for predicate prediction | Visual relationship detection · Visual relationship dataset | Predicate detection R@100: 36.15 -> 53.50 | Helped |
| Train a unified category-agnostic model rather than category-specific models (ablation) | ISIN · PASCAL VOC 2010 Part State Dataset | mAP: 8% | Hurt |
Closest papers by meaning
- Growing Interpretable Part Graphs on ConvNets via Multi-Shot Learning (AAAI 2017 · similarity 0.91)
- PARTONOMY: Large Multimodal Models with Part-Level Visual Understanding (NeurIPS 2025 · similarity 0.88)
- Multi-Part Object Representations via Graph Structures and Co-Part Discovery (NeurIPS 2026 · similarity 0.88)
- PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding (NeurIPS 2025 · similarity 0.87)
- Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation (CVPR 2025 · similarity 0.87)
- Grounding Everything: Emerging Localization Properties in Vision-Language Transformers (CVPR 2024 · similarity 0.87)
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.