The AI4Forensics Lab centers its research on developing new AI techniques for multimodal data in a forensically sound way.
To drive these advancements, the lab leverages key technical AI components, focusing primarily on:
With a commitment to explainable AI and contributing directly to a safer Dutch society through fair, unbiased, and transparent systems.
The AI4Forensics Lab is dedicated to bridging the gap between advanced multimodal AI research and forensically sound, real-world law enforcement practice.
The lab’s mission and vision are to build the fundamental knowledge base, technological tools, and human capacities necessary to transition to a new generation of criminal justice capabilities supported and enhanced by fair, probabilistic, and explainable AI systems.
The impact of the lab lies in accelerating a comprehensive sociotechnical transition towards a safer society, achieved by validating breakthroughs directly in active casework alongside the Netherlands Forensic Institute, fostering specialized scientific talent across academic and professional domains, and embedding transparent, interpretable, and accountable AI frameworks to build trust across the legal and forensic ecosystem.
Forensic detection of activities in heterogeneous data – Focuses on developing forensically sound, multimodal methods to determine a suspect’s activities by analyzing data from mobile devices and wearables (e.g., steps, heart rate, geolocations).
Forensic detection of deep fake videos – Focuses on leveraging computer vision and interactive expert visualizations to stay ahead of deep fake generation and verify whether video content is real or manipulated.
Forensic analysis of money streams in subversive crime – Focuses on using multi-scale Graph Convolutional Networks (GCNs) and interactive visualizations to detect and explain criminal patterns within complex financial transaction networks.
Forensic analysis of signals for automatic speaker comparison – Focuses on inferring the identity and topic of discussion in intercepted communications by analyzing speech content, word choice, grammar, and acoustic characteristics.
Forensic detection of steganography in images – Focuses on applying deep learning to detect covert messages concealed within digital images and uncover their hidden content.
Interpol review of detection of AI-generated image and video deepfakes, 2022-2025 Journal Article
In: vol. 13, pp. 100704, 2026.
Interactive hypergraph visual analytics for exploring large and complex image collections Journal Article
In: pp. 14738716261459480, 2026.
Forensic activity classification using digital traces from iPhones: A machine learning-based approach Journal Article
In: vol. 56, pp. 302047, 2026.
Probing Content and Channel in Speaker Verification Models Proceedings Article
In: ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026.
Interpol review of forensic image and video analysis, 2022-2025 Journal Article
In: vol. 13, pp. 100697, 2026.
De invloed van AI op forensisch bewijs in strafzaken: kansen en bedreigingen Journal Article
In: vol. 2025, no. 5, pp. 127–137, 2025.
Navigating the digital frontier–Key themes in digital forensics Journal Article
In: vol. 53, pp. 301940, 2025.
Post-clustering merging with novel metrics for multi-label image collections Journal Article
In: vol. 288, pp. 127875, 2025.
Hi-OSCAR: Hierarchical Open-set Classifier for Human Activity Recognition Journal Article
In: 2025.
REVEAL: A large-scale comprehensive image dataset for steganalysis Journal Article
In: vol. 55, pp. 302006, 2025.
Latte: Latent trajectory embedding for diffusion-generated image detection Journal Article
In: 2025.
Image Steganography Approaches and Their Detection Strategies: A Survey Journal Article
In: vol. 57, no. 2, pp. 1–40, 2025.
PaSSw0rdVib3s!: AI-assisted password recognition for digital forensic investigations Journal Article
In: vol. 52, pp. 301870, 2025.
Forensic science: an introduction to scientific and investigative techniques Book
CRC Press, 2025.
I know you’re a fraud: Uncovering illicit activity in a Greek bank transactions with unsupervised learning Journal Article
In: vol. 288, pp. 128148, 2025.
Scraping the Shadows: Deep Learning Breakthroughs in Dark Web Intelligence Journal Article
In: 2025.
Exploiting RPMB authentication in a closed source TEE implementation Journal Article
In: vol. 48, pp. 301682, 2024.
Fusing linguistic and acoustic information for automated forensic speaker comparison Journal Article
In: vol. 64, no. 5, pp. 485–497, 2024.
Improved likelihood ratios for face recognition in surveillance video by multimodal feature pairing Journal Article
In: vol. 8, pp. 100458, 2024.
Decoding Deception Journal Article
In: 2024.
Decoding deception: Understanding human discrimination ability in differentiating authentic faces from deepfake deceits Proceedings Article
In: International Conference on Image Analysis and Processing, pp. 470–481, 2023.
Forensic video steganalysis in spatial domain by noise residual convolutional neural network Journal Article
In: 2023.
Assessing data remnants in modern smartphones after factory reset Journal Article
In: vol. 46, pp. 301587, 2023.
Multi-task explainable quality networks for large-scale Forensic Facial Recognition Journal Article
In: 2023.
Frequent-words analysis for forensic speaker comparison Journal Article
In: 2023.
The influence of compression on the detection of deepfake videos Book Section
In: Artificial Intelligence (AI) in Forensic Sciences, pp. 174, 2023.
Artificial intelligence (AI) in forensic sciences Book
John Wiley & Sons, 2023.
Servicing Digital Investigations with Artificial Intelligence Book Section
In: Artificial Intelligence (AI) in Forensic Sciences, pp. 103, 2023.
Improved likelihood ratios for surveillance video face recognition with multimodal feature pairing Proceedings Article
In: 2023 11th International Workshop on Biometrics and Forensics (IWBF), pp. 1–6, 2023.
Calibration of score based likelihood ratio estimation in automated forensic facial image comparison Journal Article
In: 2022.
Likelihood ratios for deep neural networks in face comparison Journal Article
In: 2020.
Google timeline accuracy assessment and error prediction Journal Article
In: 2018.
Netherlands Forensic Institute (NFI) is one of the world’s leading forensic laboratories. It is the national forensics institute of the Netherlands. It is an autonomous division of the Dutch Ministry of Security and Justice and falls under the Directorate-General for the Administration of Justice and Law Enforcement.
University of Amsterdam (UvA) is the Netherlands’ largest university, offering the widest range of academic programmes.
