Research map / Detection and segmentation / CNN features
Frequency-Spatial Entanglement Learning for Camouflaged Object Detection
Yanguang Sun, Chunyan Xu, Jian Yang, Hanyu Xuan, Lei Luo
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 Entanglement Transformer Block (ETB) to baseline (ablation) | FSEL ResNet50-FPN baseline · CAMO | MAE: 0.093 -> 0.076 | Improved |
| add Dual-domain Reverse Parser (DRP) to baseline (ablation) | FSEL baseline · CAMO | MAE: 0.093 -> 0.074 | Improved |
| add Joint Domain Perception Module (JDPM) to baseline (ablation) | FSEL baseline · CAMO | MAE: 0.093 -> 0.081 | Improved |
| combine ETB, DRP and JDPM (ablation) | FSEL · CAMO | MAE: 0.074 -> 0.067 | Improved |
| combine all three modules (ablation) | FSEL · COD10K | S-measure: 0.778 -> 0.830 | Improved |
| use frequency-only ETB instead of spatial-only ETB (ablation) | ETB · CAMO | MAE: 0.079 -> 0.075 | Improved |
| use full entangled ETB instead of frequency-only ETB (ablation) | ETB · CAMO | MAE: 0.075 -> 0.076 | No clear effect |
| use FSEL (Ours-R50) instead of FEDER | COD · CAMO, COD10K, NC4K | MAE: 0.071 -> 0.067; 0.032 -> 0.031; 0.044 -> 0.042 | Improved |
| use FSEL PVTv2 variant instead of FSPNet | COD · COD10K | MAE: 0.026 -> 0.021 | Improved |
Closest papers by meaning
- Detecting Camouflaged Object in Frequency Domain (CVPR 2022 · similarity 0.87)
- Self-Disentanglement and Re-Composition for Cross-Domain Few-Shot Segmentation (ICML 2025 · similarity 0.85)
- FD2-Net: Frequency-Driven Feature Decomposition Network for Infrared-Visible Object Detection (AAAI 2025 · similarity 0.85)
- SdalsNet: Self-Distilled Attention Localization and Shift Network for Unsupervised Camouflaged Object Detection (AAAI 2025 · similarity 0.83)
- Endow SAM with Keen Eyes: Temporal-spatial Prompt Learning for Video Camouflaged Object Detection (CVPR 2024 · similarity 0.83)
- Triple Feature Disentanglement for One-Stage Adaptive Object Detection (AAAI 2024 · similarity 0.83)
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.
