# VAAS: Vision-Attention Anomaly Scoring for Image Manipulation Detection in Digital Forensics

arXiv: https://arxiv.org/abs/2512.15512 · HTML: https://mlsanity.com/paper/2512.15512/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 37 references that are in the map.
- First posted on arXiv 17 Dec 2025.

## Builds on (most cited first, 15 of 15)

- [CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification](https://mlsanity.com/paper/2103.14899/) (2021)
- [Celeb-DF: A Large-scale Challenging Dataset for DeepFake Forensics](https://mlsanity.com/paper/1909.12962/) (2019)
- [Media Forensics and DeepFakes: an overview](https://mlsanity.com/paper/2001.06564/) (2020)
- [Learning Rich Features for Image Manipulation Detection](https://mlsanity.com/paper/1805.04953/) (2018)
- [Hierarchical Fine-Grained Image Forgery Detection and Localization](https://mlsanity.com/paper/2303.17111/) (2023)
- [ObjectFormer for Image Manipulation Detection and Localization](https://mlsanity.com/paper/2203.14681/) (2022)
- [Hybrid LSTM and Encoder-Decoder Architecture for Detection of Image Forgeries](https://mlsanity.com/paper/1903.02495/) (2019)
- [IML-ViT: Benchmarking Image Manipulation Localization by Vision Transformer](https://mlsanity.com/paper/2307.14863/) (2023)
- [TransForensics: Image Forgery Localization with Dense Self-Attention](https://mlsanity.com/paper/2108.03871/) (2021)
- [Hotels-50K: A Global Hotel Recognition Dataset](https://mlsanity.com/paper/1901.11397/) (2019)
- [Deep Detection for Face Manipulation](https://mlsanity.com/paper/2009.05934/) (2020)
- [SoK: Exploring the State of the Art and the Future Potential of Artificial Intelligence in Digital Forensic Investigation](https://mlsanity.com/paper/2012.01987/) (2020)
- [DF-Net: The Digital Forensics Network for Image Forgery Detection](https://mlsanity.com/paper/2503.22398/) (2025)
- [DF2023: The Digital Forensics 2023 Dataset for Image Forgery Detection](https://mlsanity.com/paper/2503.22417/) (2025)
- [A Visually Attentive Splice Localization Network with Multi-Domain Feature Extractor and Multi-Receptive Field Upsampler](https://mlsanity.com/paper/2401.06995/) (2024)

## Abstract (arXiv)

Recent advances in AI-driven image generation have introduced new challenges for verifying the authenticity of digital evidence in forensic investigations. Modern generative models can produce visually consistent forgeries that evade traditional detectors based on pixel or compression artefacts. Most existing approaches also lack an explicit measure of anomaly intensity, which limits their ability to quantify the severity of manipulation. This paper introduces Vision-Attention Anomaly Scoring (VAAS), a novel dual-module framework that integrates global attention-based anomaly estimation using Vision Transformers (ViT) with patch-level self-consistency scoring derived from SegFormer embeddings. The hybrid formulation provides a continuous and interpretable anomaly score that reflects both the location and degree of manipulation. Evaluations on the DF2023 and CASIA v2.0 datasets demonstrate that VAAS achieves competitive F1 and IoU performance, while enhancing visual explainability through attention-guided anomaly maps. The framework bridges quantitative detection with human-understandable reasoning, supporting transparent and reliable image integrity assessment. The source code for all experiments and corresponding materials for reproducing the results are available open source.