This article details NIJ's recently funded research that compares deep neural network algorithms designed for face detection or recognition with other, older products.
The lack of apples-to-apples evaluations of face-detection and face-recognition algorithms has left some members of law enforcement and the security industry uncertain of what will work best for certain purposes. To help remedy this issue, the National Institute of Justice recently funded research that compared deep neural network algorithms designed for face detection or recognition with other, older products. The latter included PittPatt, the conventional algorithm long widely used by law enforcement for detection and recognition functions. Evaluations were performed by the National Criminal Justice Technology Research, Test, and Evaluation Center at the Johns Hopkins University Applied Physics Laboratory. This article discusses the test results that indicated high-powered, deep neural networks can deliver improved performance.
Downloads
Similar Publications
- A Multi-Stream Fusion Approach with One-Class Learning for Audio-Visual Deepfake Detection
- Detecting and Processing Clandestine Human Remains with Unmanned Aerial Systems and Multispectral Remote Sensing
- Development and Evaluation of a Nontargeted Electrochemical Surface-Enhanced Raman Spectroscopy (EC-SERS) Screening Method Applied to Authentic Forensic Seized Drug Casework Samples