Research map / Detection and segmentation / Open-set and few-shot detection
PASR: Pattern-Aware Scene-Conditioned Reasoning for Camouflaged Object Detection
Xinyu Wang, Jintang Xue, C.-C. Jay Kuo
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 deviation suppression (alpha=0.4 instead of 0) (ablation) | PASR · NC4K | S-measure/weighted F-measure: 0.798/0.734 -> 0.817/0.760 | Improved |
| add patch-adaptive object prototypes (ablation) | PASR · NC4K | S-measure/weighted F-measure: 0.813/0.755 -> 0.817/0.760 | Improved |
| add query-self refinement (ablation) | PASR · NC4K | S-measure/weighted F-measure: 0.808/0.748 -> 0.817/0.760 | Improved |
| add BG-anchor (coarse background alignment) (ablation) | PASR · NC4K | S-measure/weighted F-measure: 0.816/0.759 -> 0.817/0.760 | Improved |
| use scene-aligned BG-anchor instead of random BG-anchor (ablation) | PASR · NC4K | S-measure/weighted F-measure: 0.815/0.757 -> 0.817/0.760 | Improved |
| add offline SAM refinement in reference library construction (ablation) | PASR · NC4K | S-measure/weighted F-measure: 0.812/0.752 -> 0.817/0.760 | No clear effect |
| add background-conditioned deviation (alpha=0.4 instead of 0) (ablation) | PASR · CAMO | S-measure/weighted F-measure: 0.741/0.674 -> 0.755/0.693 | Improved |
| use web reference images (R1) instead of COD dataset references | PASR · CHAMELEON | S-measure: 0.832 -> 0.816 | No clear effect |
| add three-point annotations during prototype construction | PASR · COD10K | S-measure: 0.774 -> 0.787 | Improved |
| increase retrieval size K beyond 32 (ablation) | PASR · COD benchmarks | S-measure | No clear effect |
Closest papers by meaning
- Uncertainty-Guided Transformer Reasoning for Camouflaged Object Detection (ICCV 2021 · similarity 0.90)
- Shift the Lens: Environment-Aware Unsupervised Camouflaged Object Detection (CVPR 2025 · similarity 0.90)
- SAM-COD: SAM-guided Unified Framework for Weakly-Supervised Camouflaged Object Detection (ECCV 2024 · similarity 0.89)
- Beyond Weak Supervision: MLLMs-Guided Graded Knowledge Distillation for Unsupervised Camouflaged Object Detection (CVPR 2026 · similarity 0.88)
- Rethinking Detecting Salient and Camouflaged Objects in Unconstrained Scenes (ICCV 2025 · similarity 0.88)
- CamoDiffusion: Camouflaged Object Detection via Conditional Diffusion Models (AAAI 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.
