Research map / Detection and segmentation / Open-set and few-shot detection
Seeing the Unseen: A Semantic Alignment and Context-Aware Prompt Framework for Open-Vocabulary Camouflaged Object Segmentation
Peng Ren, Tian Bai, Jing Sun, Fuming Sun
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 CAP prompts to baseline (ablation) | Baseline CLIP-ConvNeXt-L · OVCamo ablation | cSm: 0.546 -> 0.623 | Improved |
| add CAFS to baseline+CAP (ablation) | Ours · OVCamo ablation | cIoU: 0.415 -> 0.506 | Improved |
| add TQD to baseline+CAP+CAFS (ablation) | Ours · OVCamo ablation | cIoU: 0.506 -> 0.540 | Improved |
| use CAP instead of handcrafted CLIP templates (ablation) | Ours · OVCamo ablation | cIoU: 0.500 -> 0.540 | Improved |
| use CAP instead of OVCoser templates (ablation) | Ours · OVCamo ablation | cSm: 0.651 -> 0.667 | Improved |
| use CAP instead of CoOp (ablation) | Ours · OVCamo ablation | cIoU: 0.513 -> 0.540 | Improved |
| use CAP instead of CoCoOp (ablation) | Ours · OVCamo ablation | cIoU: 0.532 -> 0.540 | Improved |
| extract visual tokens from F2/F3-level feature rather than global F4v (ablation) | Ours · OVCamo ablation | cIoU: 0.526 -> 0.540 | Improved |
| gating mechanism instead of element-wise addition in CAP (ablation) | Ours · OVCamo ablation | cSm: 0.655 -> 0.667 | Improved |
| CAFS interaction instead of simple multiplication (ablation) | Ours · OVCamo ablation | cIoU: 0.526 -> 0.540 | Improved |
| add semantic consistency loss Lsc (ablation) | Ours · OVCamo ablation | cIoU: 0.528 -> 0.540 | Improved |
| use CAFS-refined features in TQD instead of original CLIP features (ablation) | Ours · OVCamo ablation | cSm: 0.646 -> 0.667 | Improved |
Closest papers by meaning
- Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.89)
- Open-Vocabulary Segmentation with Semantic-Assisted Calibration (CVPR 2024 · similarity 0.89)
- Training-Free Open-Vocabulary Segmentation with Offline Diffusion-Augmented Prototype Generation (CVPR 2024 · similarity 0.87)
- Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Models (CVPR 2024 · similarity 0.87)
- OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts (NeurIPS 2025 · similarity 0.87)
- DPSeg: Dual-Prompt Cost Volume Learning for Open-Vocabulary Semantic Segmentation (CVPR 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.
