FB pixel

Innovatrics ports facial recognition to OEM and edge devices through SoC processors

Innovatrics ports facial recognition to OEM and edge devices through SoC processors
 

Innovatrics is introducing a new biometric solution called ‘SmartFace Embedded’ into chip processors to bring facial recognition to original equipment manufacturer (OEM) and edge devices.

SmartFace Embedded is described as a heavily customized version of Innovatrics’ facial recognition libraries. The platform is ported onto Ambarella CV22 and CV25 system-on-a-chip (SoC) processors designed for AI-based face biometrics processing on images and videos. The CV22 model can detect a face within 11ms (milliseconds) through on-camera processing, compared to 16ms on a CV25 chip, according to Innovatrics. SmartFace Embedded can enable pre-processing face detection and template extraction on OEM and edge devices as a result, the company says.

“With a long-term track record of porting our algorithm to various hardware, we are now able to pre-process video streams directly on a smart camera,” notes Michal Vilagi, SmartFace technical director at Innovatrics. The Slovakia-based company says its flagship SmartFace facial recognition platform can run on edge devices such as Nvidia Jetson or Blaize as well.

Innovatrics says the SoC and edge processing approach to facial recognition can simplify access control and public security by shaving down network bandwidth and server resources at the central site. This would enable more innovative deployment options, Innovatrics claims.

SmartFace Embedded is a continuation of the company’s efforts to grow in the edge processing field. In January 2022, Innovatrics released cascaded architecture based on edge video processing to support face biometrics for enhanced security.

Innovatrics also announced it will showcase its facial recognition technology at ISC West from March 22 to March 25 at the Venetian Expo in Las Vegas, Nevada.

Article Topics

 |   |   |   |   |   |   |   |   | 

Latest Biometrics News

 

MOSIP delves into biometric data quality considerations

Biometric data quality was in focus at MOSIP Connect 2026 in Rabat, Morocco, from policies for ensuring good enrollment practices…

 

NIST nominee pressed on AI standards, facial recognition oversight

The Senate Committee on Commerce, Science and Transportation on Thursday considered the nomination of Arvind Raman to serve as Under…

 

Trulioo’s Hal Lonas on how he applies aeronautics principles to fighting fraud

Rocket science is routinely held up as the ultimate example of a highly complex discipline. But Trulioo’s Hal Lonas found…

 

Vouched donates MCP-I framework to Decentralized Identity Foundation

An announcement from Seattle-based Vouched says it has formally donated its Model Context Protocol – Identity (MCP-I) framework to the…

 

California’s OS-based age verification law challenges open-source community

California’s new online safety bill, AB 1043 (the Digital Age Assurance Act), adopts a declared age model for operating systems….

 

87% of failed biometric verifications in Southern Africa due to AI spoofing: Smile ID

A new report spotlights deepfake fraud posing an acute problem for Africa. Digital identity, banking and e-government are being used…

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Biometric Market Analysis and Buyer's Guides

Most Viewed This Week

Featured Company

Biometrics Insight, Opinion

Digital ID In-Depth

Biometrics White Papers

Biometrics Events