Research map / Detection and segmentation / Open-set and few-shot detection
Unlocking Attributes' Contribution to Successful Camouflage: A Combined Textual and Visual Analysis Strategy
Hong Zhang, Yixuan Lyu, Qian Yu, Hanyang Liu, Huimin Ma, Yuan Ding, Yifan Yang
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 fixation prediction module (ablation) | ACUMEN baseline · COD10K | F_beta^w: 0.72491 -> 0.74769 | Improved |
| add attribute contribution prediction (ablation) | ACUMEN baseline · COD10K | F_beta^w: 0.72491 -> 0.75853 | Improved |
| add fixation on top of attribute prediction (ablation) | ACUMEN Attr · COD10K | F_beta^w: 0.75853 -> 0.75797 | No clear effect |
| add text-visual consistency loss during training (ablation) | ACUMEN Fix + Attr · COD10K | F_beta^w: 0.75797 -> 0.76149 | Improved |
| add attribute prediction (ablation) | ACUMEN · Params | params: 227.99 -> 480.61 (Params) | Hurt |
| use N=3 fixation decoder layers and M=1 mask decoder layer (ablation) | ACUMEN · CAMO | F_beta^w: 0.83766 -> 0.8496 | Improved |
| increase mask decoder layers to M=6 with N=1, WL=50 (ablation) | ACUMEN · CAMO | F_beta^w: 0.84328 (M=3) -> 0.80757 (M=6) | Hurt |
| reduce text word length from 77 to 50 (ablation) | ACUMEN · COD10K | F_beta^w: 0.7467 -> 0.7509 (N=1,M=1); 0.74543 -> 0.75671 (N=1,M=3) | Mixed |
| ACUMEN vs FSPNet on NC4K | ACUMEN · NC4K | S_alpha: 0.874 vs 0.879 | No clear effect |
Closest papers by meaning
- MMCSBench: A Fine-Grained Benchmark for Large Vision-Language Models in Camouflage Scenes (NeurIPS 2025 · similarity 0.90)
- The Making and Breaking of Camouflage (ICCV 2023 · similarity 0.89)
- Simultaneously Localize, Segment and Rank the Camouflaged Objects (CVPR 2021 · similarity 0.89)
- Camouflage Anything: Learning to Hide using Controlled Out-painting and Representation Engineering (CVPR 2025 · similarity 0.88)
- Shift the Lens: Environment-Aware Unsupervised Camouflaged Object Detection (CVPR 2025 · similarity 0.86)
- MM-CamObj: A Comprehensive Multimodal Dataset for Camouflaged Object Scenarios (AAAI 2025 · similarity 0.86)
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.
