Publication Date

Spring 2022

Degree Type

Master's Project

Degree Name

Master of Science (MS)


Computer Science

First Advisor

Teng Moh

Second Advisor

Mark Barash

Third Advisor

William Andreopoulos


DNA Face Classification


A large majority of crimes such as homicides, sexual assaults and missing person cases are not solved within a reasonable timeframe and become cold cases. The ability to predict visual appearance and ancestry from a DNA sample will provide an unprecedented advancement in such criminal investigations. DNA based prediction of craniofacial features, phenotypes and ancestry can be used to reduce the pool of candidates onto which to perform further investigations. To achieve the above goal, it is first essential to substantiate, model and measure the intrinsic relationship between the genomic markers and phenotypic features. The first step is to standardize the 3D face scans using the CoMA data format followed by its projection into a low- dimensional latent embedding space. The second step is to reduce the dimensionality of the genetic space by performing Principal Component Analysis on the genomic markers to generate compact genomic properties. A simple multi-layer perceptron is trained to classify an ensemble of facial embeddings and genomic properties into genuine and imposter pairings. The classification model is able to match the DNA with the given 3D face with an average AUC score of 0.73. The introduction of hand-picked genomic markers serves to be an important contribution towards improving the final AUC score. Furthermore, results indicate that incorporating additional phenotypical properties such as sex and age leads to a better verification. Thus, this study proves to be an important milestone towards identifying suspects from their DNA sample in a criminal investigation.