Neural Boneprint: Person Identification from Bones using Generative Contrastive Deep Learning

Chaoqun Niu, Dongdong Chen, Ji-Zhe Zhou, Jian Wang, Xiang Luo, Quan-Hui Liu, YUAN LI, Jiancheng Lv

Abstract:

Forensic person identification is of paramount importance in accidents and criminal investigations. Existing methods based on soft tissue or DNA can be unavailable if the body is badly decomposed, white-ossified, or charred. However, bones last a long time. This raises a natural question: ***can we learn to identify a person using bone data?***
We present a novel feature of bones called ***Neural Boneprint*** for personal identification. In particular, we exploit the thoracic skeletal data including chest radiographs (CXRs) and computed tomography (CT) images enhanced by the volume rendering technique (VRT) as an example to explore the availability of the neural boneprint. We then represent the neural boneprint as a joint latent embedding of VRT images and CXRs through a bidirectional cross-modality translation and contrastive learning. Preliminary experimental results on real skeletal data demonstrate the effectiveness of the Neural Boneprint for identification. We hope that this approach will provide a promising alternative for challenging forensic cases where conventional methods are limited. The code will be available at ***.

Primary Subject Area: [Content] Media Interpretation

Secondary Subject Area: [Content] Media Interpretation

Relevance To Conference: This study innovatively proposes the concept of "neural boneprint" to address the challenges of person identification in forensic scenarios. It validates this concept using multimodal thoracic skeletal data. It could potentially serve as a complementary or even alternative method to existing technologies such as DNA testing, fingerprint recognition, and facial recognition. The study particularly emphasizes the use of cross-modality translation and cross-modality fusion methods to obtain the neural boneprint, a data-driven feature representation. This effectively extracts and integrates individual identity information from skeletal images of different modalities/views for person identification, significantly improving the efficiency of large-scale data processing and reducing the reliance on expert knowledge.

Submission Number: 2869