Research map / Detection and segmentation / Open-set and few-shot detection
FocusDiffuser: Perceiving Local Disparities for Camouflaged Object Detection
Jianwei Zhao, Xin Li, Fan Yang, Qiang Zhai, Ao Luo, Zhicheng Jiao, Hong Cheng
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 |
|---|---|---|---|
| replace transformer decoder with diffusion model (ablation) | FocusDiffuser (PVTv2) · COD10K | weighted F-measure: 0.673 -> 0.787 | Improved |
| add BDLU module (ablation) | diffusion baseline · COD10K | MAE: 0.033 -> 0.027 | Improved |
| add Cyclic Positioning module (ablation) | diffusion baseline · COD10K | MAE: 0.033 -> 0.022 | Improved |
| combine BDLU and CP (ablation) | diffusion baseline · COD10K | MAE: 0.033 -> 0.020 | Improved |
| add correlation LookUp without boundary selection (ablation) | BDLU · COD10K | S-alpha: 0.869 -> 0.867 | Hurt |
| add boundary selection to lookup (ablation) | BDLU · COD10K | S-alpha: 0.867 -> 0.875 | Improved |
| increase lookup grid radius from 2 to 4 (ablation) | BDLU · COD10K | S-alpha: 0.875 -> 0.865 | Hurt |
| use bounding-box mask for background filtering in CP instead of all-ones mask (ablation) | CP · COD10K | S-alpha: 0.867 -> 0.875 | Improved |
| increase denoising steps beyond 4 (ablation) | FocusDiffuser · COD10K | S-alpha: 0.875 -> 0.875 (N=8) | No clear effect |
| reduce denoising steps from 4 to 2 (ablation) | FocusDiffuser · COD10K | S-alpha: 0.875 -> 0.870 | Hurt |
| lower signal scale factor b from 1.0 to 0.1 (ablation) | FocusDiffuser · COD10K | S-alpha: 0.859 -> 0.875 | Improved |
| lower scale factor b further from 0.1 to 0.05 (ablation) | FocusDiffuser · COD10K | S-alpha: 0.875 -> 0.870 | Hurt |
Closest papers by meaning
- CamoDiffusion: Camouflaged Object Detection via Conditional Diffusion Models (AAAI 2024 · similarity 0.91)
- Conditional Diffusion Models for Camouflaged and Salient Object Detection (TPAMI 2025 · similarity 0.91)
- CODiff: One-Step Diffusion Model for Camouflaged Object Detection (ICML 2026 · similarity 0.90)
- Shift the Lens: Environment-Aware Unsupervised Camouflaged Object Detection (CVPR 2025 · similarity 0.89)
- SdalsNet: Self-Distilled Attention Localization and Shift Network for Unsupervised Camouflaged Object Detection (AAAI 2025 · similarity 0.89)
- Camouflage Anything: Learning to Hide using Controlled Out-painting and Representation Engineering (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.
