In this session we describe our work developing software for a custom camera system based on GStreamer. The BRIAR project requires high-quality video capture at distances over 400m for biometric recognition and identification, but commercial cameras struggle to capture high-quality video at these distances. To address this, we developed a custom camera system using GStreamer, enabling advanced imaging capabilities and long-range data capture. Our work included designing a GStreamer pipeline capable of managing multiple sensor formats, integrating UDP server hooks to manage recording, using GstRTSPServer to build an RTSP server and creating an extensible hardware control interface. By integrating with network video recorders, we simplified monitoring, data handling and curation, and successfully supported BRIAR’s complex data collection efforts. This presentation will detail our GStreamer-based implementation, highlighting the technical challenges we faced and how we overcame them.
Developing a GStreamer-Based Custom Camera System for Long-Range Biometric Data Collection
Session Speakers
-
Gavin Jager
Researcher and Lab Space Manager, Oak Ridge National Laboratory
Gavin Jager is a Research Assistant in the Human Analysis and Biometrics Group at Oak Ridge National Laboratory. He has worked there since 2021, shortly after graduating from the University of Michigan with a BS in Physics and German. His focus so far has been biometric dataset collection and curation, utilizing a range of commercial sensors and developing custom camera systems and software. Gavin is a major contributor to the collection and curation efforts that produced the BRIAR datasets. He is currently working on imaging, biometric signals and time series classification projects.
