Research map / Image generation / Autoregressive and latent generation
BioFors: A Large Biomedical Image Forensics Dataset
Ekraam Sabir, Soumyaroop Nandi, Wael Abd-Almageed, Prem Natarajan
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 |
|---|---|---|---|
| use dense feature matching (DenseField ZM) instead of sparse keypoints (SIFT) | external duplication detection · BioFors EDD combined | MCC image-level: 0.142 -> 0.278 | Improved |
| use dense matching (DF-PCT) instead of block-based DCT | internal duplication detection · BioFors IDD combined | MCC image-level: 0.196 -> 0.569 | Improved |
| use DF-PCT instead of deep BusterNet | internal duplication detection · BioFors IDD combined | MCC pixel-level: 0.107 -> 0.364 | Improved |
| use deep DMVN instead of DF-ZM | external duplication detection · BioFors EDD Blot/Gel | MCC pixel-level: 0.192 -> 0.430 | Mixed |
| use fine-tuned ManTraNet instead of a simple CNN baseline | cut/sharp-transition detection · BioFors CSTD | image-level F1: 0.212 -> 0.253 | Improved |
| use ManTraNet instead of CNN baseline | cut/sharp-transition detection · BioFors CSTD | image-level MCC: 0.098 -> 0.170 | Improved |
| use DenseNet instead of VGG16 for image-class classification | BioFors image classification · BioFors test | accuracy: 97.11% -> 97.67% | No clear effect |
Closest papers by meaning
- Rescind: Countering Image Misconduct in Biomedical Publications with Vision-Language and State-Space Modeling (AAAI 2026 · similarity 0.87)
- ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and Localization (NeurIPS 2025 · similarity 0.83)
- BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature (CVPR 2025 · similarity 0.83)
- Community Forensics: Using Thousands of Generators to Train Fake Image Detectors (CVPR 2025 · similarity 0.83)
- THEMIS: Towards Holistic Evaluation of MLLMs for Scientific Paper Fraud Forensics (ICLR 2026 · similarity 0.82)
- Forensic Self-Descriptions Are All You Need for Zero-Shot Detection, Open-Set Source Attribution, and Clustering of AI-generated Images (CVPR 2025 · similarity 0.82)
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.
