FCR: Investigating Generative AI models for Forensic Craniofacial Reconstruction

2025-08-25

AI Summary

Forensic craniofacial reconstruction faces two key bottlenecks: the labor-intensive and time-consuming nature of traditional manual approaches, and the inability of existing generative models to effectively capture the cross-domain mapping between skull morphology and facial appearance. To address these challenges, this paper proposes the first end-to-end cross-domain generation framework for facial reconstruction from 2D X-ray skull images. Our method innovatively integrates CycleGAN and conditional GAN architectures, incorporates anatomical constraints, employs fine-grained discriminator optimization, and introduces a generated-image retrieval module for identity matching. Quantitative evaluation demonstrates significant improvements over baseline models across standard metrics—including FID, Inception Score (IS), and SSIM—yielding photorealistic facial details and high anatomical fidelity. Retrieval experiments further confirm robust alignment with real-face databases. This work establishes an efficient, deployable AI paradigm for forensic identification.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Web data generation and simulation
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Craniofacial reconstruction in forensics is one of the processes to identify victims of crime and natural disasters. Identifying an individual from their remains plays a crucial role when all other identification methods fail. Traditional methods for this task, such as clay-based craniofacial reconstruction, require expert domain knowledge and are a time-consuming process. At the same time, other probabilistic generative models like the statistical shape model or the Basel face model fail to capture the skull and face cross-domain attributes. Looking at these limitations, we propose a generic framework for craniofacial reconstruction from 2D X-ray images. Here, we used various generative models (i.e., CycleGANs, cGANs, etc) and fine-tune the generator and discriminator parts to generate more realistic images in two distinct domains, which are the skull and face of an individual. This is the first time where 2D X-rays are being used as a representation of the skull by generative models for craniofacial reconstruction. We have evaluated the quality of generated faces using FID, IS, and SSIM scores. Finally, we have proposed a retrieval framework where the query is the generated face image and the gallery is the database of real faces. By experimental results, we have found that this can be an effective tool for forensic science.

Problem

• Reconstructing faces from skulls for forensic identification
• Overcoming limitations of traditional clay-based reconstruction methods
• Using generative AI with 2D X-rays to create realistic facial images

Innovation

• Generative models for skull-face reconstruction
• Fine-tuning GANs on 2D X-ray images
• Novel retrieval framework using generated faces