Research map / Detection and segmentation / Open-set and few-shot detection
Iterative Refinement of Semantic and Spatial Representations for Open-Vocabulary Camouflaged Object Segmentation
Fangyan Wang, Ge Jiao, Guowen Yue
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 category re-ranker (ablation) | Baseline (CLIP + SAM2) · OVCamo (ablation) | cSm: 0.658 -> 0.678 | Improved |
| add segmentation modulator (ablation) | +R · OVCamo (ablation) | cSm: 0.678 -> 0.697 | Improved |
| add learnable prompt (ablation) | +R, M · OVCamo (ablation) | cSm: 0.697 -> 0.711 | Improved |
| add iterative refinement (ablation) | +R, M, L · OVCamo (ablation) | cSm: 0.711 -> 0.723 | Improved |
| add iterative refinement (ablation) | +R, M, L · OVCamo (ablation) | cMAE: 0.230 -> 0.211 | Improved |
| raise iterations from 1 to 3 (ablation) | ISSR · OVCamo (ablation) | cSm: 0.714 -> 0.723 | Improved |
| raise iterations from 3 to 5 (ablation) | ISSR · OVCamo (ablation) | cSm: 0.723 -> 0.711 | Hurt |
| use prompt context length 16 instead of 4 (ablation) | ISSR · OVCamo (ablation) | cSm: 0.714 -> 0.723 | Improved |
| use prompt context length 16 instead of 64 (ablation) | ISSR · OVCamo (ablation) | cSm: 0.713 -> 0.723 | Improved |
| raise Top-N from 2 to 10 (ablation) | ISSR · OVCamo (ablation) | cSm: 0.703 -> 0.723 | Improved |
| raise Top-N from 10 to 14 (ablation) | ISSR · OVCamo (ablation) | cSm: 0.723 -> 0.714 | Hurt |
| add image visual features to re-ranker input (ablation) | re-ranker with S only · OVCamo (ablation) | cSm: 0.711 -> 0.715 | Improved |
Closest papers by meaning
- Training-Free Open-Vocabulary Camouflaged Object Segmentation via Fine-Grained Object Binding and Adaptive Hybrid Prompt (CVPR 2026 · similarity 0.91)
- Open-Vocabulary Camouflaged Object Segmentation (ECCV 2024 · similarity 0.91)
- OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts (NeurIPS 2025 · similarity 0.88)
- Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.87)
- CamSAM2: Segment Anything Accurately in Camouflaged Videos (NeurIPS 2025 · similarity 0.87)
- SANSA: Unleashing the Hidden Semantics in SAM2 for Few-Shot Segmentation (NeurIPS 2025 · 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.
