A Novel Evolutionary Method for Automated Skull-Face Overlay in Computer-Aided Craniofacial Superimposition
This method offers improved accuracy and robustness for forensic practitioners identifying skeletal remains, addressing a critical bottleneck in craniofacial superimposition.
The paper addresses the challenge of Skull-Face Overlay (SFO) in forensic craniofacial superimposition, where aligning a 3D skull with a 2D facial image is complicated by soft-tissue thickness variability. Lilium, an automated evolutionary method, explicitly models this variability using a 3D cone-based representation optimized by Differential Evolution, outperforming state-of-the-art methods in accuracy and robustness.
Craniofacial Superimposition is a forensic technique for identifying skeletal remains by comparing a post-mortem skull with ante-mortem facial photographs. A critical step in this process is Skull-Face Overlay (SFO). This stage involves aligning a 3D skull model with a 2D facial image, typically guided by cranial and facial landmarks' correspondence. However, its accuracy is undermined by individual variability in soft-tissue thickness, introducing significant uncertainty into the overlay. This paper introduces Lilium, an automated evolutionary method to enhance the accuracy and robustness of SFO. Lilium explicitly models soft-tissue variability using a 3D cone-based representation whose parameters are optimized via a Differential Evolution algorithm. The method enforces anatomical, morphological, and photographic plausibility through a combination of constraints: landmark matching, camera parameter consistency, head pose alignment, skull containment within facial boundaries, and region parallelism. This emulation of the usual forensic practitioners' approach leads Lilium to outperform the state-of-the-art method in terms of both accuracy and robustness.