Novetta whitepaper analyzes Amazon facial recognition service
Novetta has released a new whitepaper titled “Amazon Rekognition: Quick-Look Biometric Performance Assessment“, which analyzes the biometric performance of Amazon’s machine learning-based face recognition service.
With its relatively low cost, ease of implementation, and use of deep learning, Amazon Rekognition has gained significant traction in government and commercial applications.
The whitepaper Novetta provides several new insights into Amazon Rekognition performance.
For instance, the machine learning-based facial recognition service is able to correctly identified about 90 percent of subjects against relatively small databases.
The paper also confirmed that Rekognition enrollment error rates were approximately 10-times lower than those of other cloud-based face recognition tools.
Finally, Novetta analysts determined that Rekognition is calibrated to reduce false positive errors, even if this leads to relatively high false negative errors.
“Amazon Rekognition – along with other machine learning-based approaches – is emerging as a disruptive capability in US government applications,” said Michael Thieme, Novetta’s vice president of special projects at Novetta. “Understanding its real-world performance is a precondition of effective use in surveillance, large-scale identification, and social media applications.”
Last month, Chris Adzima, a senior information systems analyst for Washington County Sheriff’s Office discussed how his team uses AWS for facial recognition as well as Amazon Rekognition as a crime-solving tool to identify persons of interest.