Research map / Detection and segmentation / CNN features
Image Manipulation Detection by Multi-View Multi-Scale Supervision
Xinru Chen, Chengbo Dong, Jiaqi Ji, Juan Cao, Xirong Li
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 image-scale classification loss to Seg (ablation) | Seg · DEFACTO-12k | pixel-level mean F1: 0.546 -> 0.463 | Hurt |
| add image-scale classification loss to Seg (ablation) | Seg · DEFACTO-12k | image-level specificity: 0.620 -> 0.778 | Improved |
| add image-scale classification loss to Seg (ablation) | Seg · DEFACTO-12k | Com-F1: 0.617 -> 0.579 | Hurt |
| add noise-sensitive branch (NSB) to Seg+Clf (ablation) | Seg+Clf · DEFACTO-12k | Com-F1: 0.579 -> 0.619 | Improved |
| add edge-supervised branch (ESB) to Seg+Clf (ablation) | Seg+Clf · DEFACTO-12k | Com-F1: 0.579 -> 0.626 | Improved |
| add Sobel layer to ESB (ablation) | Seg+Clf+E · DEFACTO-12k | copy-move pixel F1: 0.382 -> 0.405 | Improved |
| use ESB instead of GSR-Net edge branch (ablation) | Seg+Clf · DEFACTO-12k | Com-F1: 0.613 -> 0.626 | Improved |
| fuse ESB and NSB with dual attention instead of model averaging (ablation) | MVSS-Net · DEFACTO-12k | Com-F1: 0.622 -> 0.643 | Improved |
| combine ESB and NSB (full setup) versus ESB alone (ablation) | MVSS-Net · DEFACTO-12k | Com-F1: 0.626 -> 0.643 | Improved |
| full multi-view multi-scale MVSS-Net versus plain FCN, fixed threshold | MVSS-Net · 5 test sets (trained on CASIAv2) | pixel-level mean F1: 0.232 -> 0.394 | Improved |
| MVSS-Net versus FCN on inpainting subsets: optimal threshold mean | MVSS-Net · 5 test sets (trained on CASIAv2) | pixel-level mean F1: 0.562 -> 0.718 | Improved |
| MVSS-Net versus HP-FCN on inpainting subset of NIST16 | MVSS-Net · NIST16 inpainting | pixel-level F1: 0.284 -> 0.565 | 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.
- ManTra-Net — ManTra-Net: Manipulation Tracing Network for Detection and Localization of Image Forgeries With Anomalous Features (CVPR 2019) · better on 2 of 2 shared results
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
- UnionFormer: Unified-Learning Transformer with Multi-View Representation for Image Manipulation Detection and Localization (CVPR 2024 · similarity 0.86)
- IMDPrompter: Adapting SAM to Image Manipulation Detection by Cross-View Automated Prompt Learning (ICLR 2025 · similarity 0.86)
- IDseq: Decoupled and Sequentially Detecting and Grounding Multi-Modal Media Manipulation (AAAI 2025 · similarity 0.85)
- ForgerySleuth: Empowering Multimodal Large Language Models for Image Manipulation Detection (NeurIPS 2025 · similarity 0.85)
- A New Benchmark and Model for Challenging Image Manipulation Detection (AAAI 2024 · similarity 0.85)
- Towards Modern Image Manipulation Localization: A Large-Scale Dataset and Novel Methods (CVPR 2024 · similarity 0.85)
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.
