Research map / Detection and segmentation / CNN features
VCP-DCN: Beyond Visual Concealed Property via Depth Collaborative Network for Camouflaged Object Detection
Songsong Duan, Xi Yang, Nannan Wang
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 the SPE module to the RGB-D baseline (ablation) | VCP-DCN baseline · CAMO | MAE: 0.052 -> 0.045 | Improved |
| add MDA on top of SPE (ablation) | VCP-DCN (SPE) · CAMO | MAE: 0.045 -> 0.039 | Improved |
| add DAI on top of SPE (ablation) | VCP-DCN (SPE) · CAMO | MAE: 0.045 -> 0.041 | Improved |
| add DAI on top of SPE and MDA (ablation) | VCP-DCN (SPE+MDA) · CAMO | MAE: 0.039 -> 0.033 | Improved |
| add foreground masked linear attention to MDA-less model (ablation) | VCP-DCN (A1) · CAMO | MAE: 0.041 -> 0.037 | Improved |
| add background masked linear attention to MDA-less model (ablation) | VCP-DCN (A1) · CAMO | MAE: 0.041 -> 0.039 | Improved |
| use both foreground and background MLA instead of foreground only (ablation) | VCP-DCN MDA · CAMO | MAE: 0.037 -> 0.033 | Improved |
| add modality-specific features in DAI (ablation) | VCP-DCN (B1, no DAI) · CAMO | MAE: 0.039 -> 0.035 | Improved |
| add modality-consistency features in DAI (ablation) | VCP-DCN (B1, no DAI) · CAMO | MAE: 0.039 -> 0.038 | Improved |
| use Depth Anything V2 depth instead of DPT depth (ablation) | VCP-DCN (VMamba) · CAMO | S-measure: 0.875 -> 0.892 | Improved |
| use Depth Anything V2 depth instead of DPT depth (ablation) | VCP-DCN (VMamba) · COD10K | S-measure: 0.862 -> 0.881 | Improved |
| use VCP-DCN instead of Samba with same backbone and depth | Samba (VMamba, DAMv2) · CAMO | MAE: 0.036 -> 0.033 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SAM-COD — SAM-COD: SAM-guided Unified Framework for Weakly-Supervised Camouflaged Object Detection (ECCV 2024) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ZoomNeXt: A Unified Collaborative Pyramid Network for Camouflaged Object Detection (TPAMI 2024 · similarity 0.90)
- Zoom in and Out: A Mixed-Scale Triplet Network for Camouflaged Object Detection (CVPR 2022 · similarity 0.90)
- ESCNet:Edge-Semantic Collaborative Network for Camouflaged Object Detection (ICCV 2025 · similarity 0.89)
- Camouflaged Object Detection (CVPR 2020 · similarity 0.89)
- Shift the Lens: Environment-Aware Unsupervised Camouflaged Object Detection (CVPR 2025 · similarity 0.88)
- SdalsNet: Self-Distilled Attention Localization and Shift Network for Unsupervised Camouflaged Object Detection (AAAI 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.
