Hand-Based Person Identification using Global and Part-Aware Deep Feature Representation Learning
This work addresses a critical forensic identification challenge for law enforcement, where hand images are often the only available evidence in serious crimes like sexual abuse.
The paper tackles the problem of person identification from hand images in uncontrolled crime scenes by proposing a deep learning method that learns both global and local feature representations, achieving significant performance improvements over competing approaches on two large multi-ethnic datasets.
In cases of serious crime, including sexual abuse, often the only available information with demonstrated potential for identification is images of the hands. Since this evidence is captured in uncontrolled situations, it is difficult to analyse. As global approaches to feature comparison are limited in this case, it is important to extend to consider local information. In this work, we propose hand-based person identification by learning both global and local deep feature representations. Our proposed method, Global and Part-Aware Network (GPA-Net), creates global and local branches on the conv-layer for learning robust discriminative global and part-level features. For learning the local (part-level) features, we perform uniform partitioning on the conv-layer in both horizontal and vertical directions. We retrieve the parts by conducting a soft partition without explicitly partitioning the images or requiring external cues such as pose estimation. We make extensive evaluations on two large multi-ethnic and publicly available hand datasets, demonstrating that our proposed method significantly outperforms competing approaches.