Research map / Image generation / GANs
Reality Transform Adversarial Generators for Image Splicing Forgery Detection and Localization
Xiuli Bi, Zhipeng Zhang, Bin Xiao
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 fake-to-realistic transform generator G_T with adversarial training (ablation) | RTAG localization generator G_M · CASIA v2.0 | mAP / F: mAP 0.606 -> 0.707; F 0.704 -> 0.818 | Improved |
| use multi-decoder-single-task instead of single decoder for all tasks (ablation) | G_M · CASIA v2.0 | mAP / F: mAP 0.622 -> 0.707; F 0.731 -> 0.818 | Improved |
| add reconstruction decoder (ablation) | RTAG · CASIA v2.0 | mAP / F: mAP 0.684 -> 0.707; F 0.772 -> 0.818 | Improved |
| add edge decoder (ablation) | RTAG · CASIA v2.0 | mAP / F: mAP 0.689 -> 0.707; F 0.775 -> 0.818 | Improved |
| learn alpha in WCT blocks instead of fixing it manually (1, 0.5, 0.2) (ablation) | G_T · qualitative | visual quality | Improved |
| use RTAG instead of best prior method C2RNet | splicing localization · NIST 2016 | mAP: 0.097 -> 0.531 | Improved |
| use RTAG instead of MAG | splicing localization · FantasticReality | mAP: 0.780 -> 0.910 | Improved |
Closest papers by meaning
- DOA-GAN: Dual-Order Attentive Generative Adversarial Network for Image Copy-Move Forgery Detection and Localization (CVPR 2020 · similarity 0.88)
- Active Adversarial Noise Suppression for Image Forgery Localization (TPAMI 2026 · similarity 0.87)
- The Point Where Reality Meets Fantasy: Mixed Adversarial Generators for Image Splice Detection (NeurIPS 2019 · similarity 0.87)
- Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection (CVPR 2024 · similarity 0.87)
- FakeShield: Explainable Image Forgery Detection and Localization via Multi-modal Large Language Models (ICLR 2025 · similarity 0.86)
- Learning Discriminative Noise Guidance for Image Forgery Detection and Localization (AAAI 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.
