BrainChip increases facial classification accuracy up to 30 percent

Dec 21, 2018 | Stephen Mayhew

BrainChip has announced an update of its AI-powered video analysis software BrainChip Studio that it says improves its face classification accuracy by between and 10 and 30 percent.

The new BrainChip Studio 2018.3 introduces a full-face mode to performs facial classifications, in addition to the partial-face mode enabled by spiking neural networks in previous versions. The partial-face mode is effective for matching images with faces obscured by hats, masks, camera angles, or other obstacles. In situations in which the probe image or extracted face is a complete facial image, the accuracy of the new version is improved without affecting throughput, according to the announcement.

“We are always looking for ways to continually improve our products by listening to our customer requests,” said Bob Beachler, BrainChip’s Senior Vice President of Marketing and Business Development. “Not surprisingly, improving accuracy is typically at the top of list for video analytic software. With BrainChip Studio 2018.3 we were able to provide a dramatic increase in accuracy.”

The company cites a forecast by MarketsandMarkets that the facial recognition market will surpass $7 billion annually by 2022, and says that its technology works in environments where low light, low resolution, or visual noise would defeat other biometric-based facial recognition systems.

BrainChip recently formed a strategic partnership with Telesikring to address the law enforcement and enterprise security markets in Scandanavia.