Research map / Detection and segmentation / Open-set and few-shot detection
Focus Manipulation Detection via Photometric Histogram Analysis
Can Chen, Scott McCloskey, Jingyi Yu
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 |
|---|---|---|---|
| combine five FMIH feature classifiers by majority voting | FMIHNet · FMD | classification accuracy: 0.942 -> 0.982 | Improved |
| represent features as intensity-versus-feature FMIH histograms rather than image pixels | FMD image classification · FMD | classification accuracy: 0.784 -> 0.982 | Improved |
| add the NOI FMIH feature to the full feature ensemble (ablation) | FMIHNet · FMD | classification accuracy: 0.967 -> 0.982 | Improved |
| add the GRAD FMIH feature to the full feature ensemble (ablation) | FMIHNet · FMD | classification accuracy: 0.954 -> 0.982 | Improved |
| add the ADQ FMIH feature to the full feature ensemble (ablation) | FMIHNet · FMD | classification accuracy: 0.946 -> 0.982 | Improved |
| add the CFA FMIH feature to the full feature ensemble (ablation) | FMIHNet · FMD | classification accuracy: 0.919 -> 0.982 | Improved |
| add the VAR FMIH feature to the full feature ensemble (ablation) | FMIHNet · FMD | classification accuracy: 0.850 -> 0.982 | Improved |
| train on combined FMD and NC2017 examples instead of NC2017 only (ablation) | FMIHNet vote · NC2017 | classification accuracy: 0.630 -> 0.771 | Improved |
| train on combined FMD and NC2017 examples instead of FMD only (ablation) | FMIHNet vote · NC2017 | classification accuracy: 0.623 -> 0.771 | Improved |
| use intensity-feature histograms rather than feature maps for NOI classification | NOI classifier · NC2017 | classification accuracy: 0.667 -> 0.760 | Improved |
Closest papers by meaning
- Fighting Fake News: Image Splice Detection via Learned Self-Consistency (ECCV 2018 · similarity 0.85)
- A New Benchmark and Model for Challenging Image Manipulation Detection (AAAI 2024 · similarity 0.84)
- BIAS-ID: A Framework for Analyzing Transformation Biases in AI-Generated Image Detectors (NeurIPS 2026 · similarity 0.83)
- Learning Discriminative Noise Guidance for Image Forgery Detection and Localization (AAAI 2024 · similarity 0.83)
- Towards Modern Image Manipulation Localization: A Large-Scale Dataset and Novel Methods (CVPR 2024 · similarity 0.83)
- ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and Localization (NeurIPS 2025 · 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.
