Research map / Detection and segmentation / CNN features
High-Resolution Iterative Feedback Network for Camouflaged Object Detection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use HitNet instead of LSR | COD10K-Test · COD10K-Test | Fβw: 0.685 -> 0.798 | Helped |
| Use HitNet instead of SINet-V2 | CAMO-Test · CAMO-Test | MAE: 0.070 -> 0.057 | Helped |
| Add multi-resolution iterative refinement (RIR) (ablation) | HitNet · COD10K | Fβw: 0.703 -> 0.798 | Helped |
| Use PVT feature extraction instead of Res2Net-50 (ablation) | HitNet · COD10K | Fβw: 0.735 -> 0.798 | Helped |
| Use IFF instead of convolutional fusion (ablation) | HitNet · COD10K | Fβw: 0.791 -> 0.798 | Helped |
| Increase input resolution from 352×352 to 704×704 (ablation) | HitNet · COD10K | Fβw: 0.727 -> 0.798 | Helped |
| Increase iteration count from 1 to 4 (ablation) | HitNet · COD10K | MAE: 0.0252 -> 0.0240 | Helped |
| Increase iteration count from 4 to 5 (ablation) | HitNet · COD10K | MAE: 0.0240 -> 0.0241 | Hurt |
| Add iterative loss supervision (Tie) (ablation) | HitNet feedback mechanism · COD10K | MAE: 0.0268 -> 0.0255 | Helped |
| Add the feedback block (FB) after Tie (ablation) | HitNet feedback mechanism · COD10K | MAE: 0.0255 -> 0.0248 | Helped |
| Add multi-scale connection fusion after Tie and FB (ablation) | HitNet feedback mechanism · COD10K | MAE: 0.0248 -> 0.0240 | Helped |
| Train with salient data without cross-domain learning (ablation) | HitNet · COD10K | Fβw: 0.798 -> 0.782 | Hurt |
Closest papers by meaning
- Zoom in and Out: A Mixed-Scale Triplet Network for Camouflaged Object Detection (CVPR 2022 · similarity 0.89)
- Segment, Magnify and Reiterate: Detecting Camouflaged Objects the Hard Way (CVPR 2022 · similarity 0.88)
- SdalsNet: Self-Distilled Attention Localization and Shift Network for Unsupervised Camouflaged Object Detection (AAAI 2025 · similarity 0.85)
- Shift the Lens: Environment-Aware Unsupervised Camouflaged Object Detection (CVPR 2025 · similarity 0.85)
- CamoDiffusion: Camouflaged Object Detection via Conditional Diffusion Models (AAAI 2024 · similarity 0.84)
- Camouflage Anything: Learning to Hide using Controlled Out-painting and Representation Engineering (CVPR 2025 · similarity 0.84)
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.