The Department of Homeland Security has tested face recognition algorithms using video data to support potential law enforcement applications.
DHS said Tuesday its science and technology directorate worked with the National Institute of Standards and Technology to independently evaluate 36 face recognition algorithms as part of the Face in Video Evaluation program.
FIVE used video sequences collected in 2013 at the Toyota Center multipurpose stadium in Kennewick, Washington, as well as video and camera positioning data from the Homeland Security Advanced Research Projects Agency’s Apex Air Entry/Exit Re-engineering initiative.
Volunteers participated in data collection efforts at the Toyota Center and simulated three operational settings including unidirectional crowd flow, bi-directional crowd flow and winding queue, DHS noted.
The department also released a report on the findings from the FIVE program.
DHS made the report to help law enforcement officials explore and buy technologies that can help address their organizations’ face detection and recognition needs.