Blaize and Innovatrics Provide Edge Ready, Low Power, Low Latency Facial Recognition Technology

By Tiera Oliver

Associate Editor

Embedded Computing Design

May 16, 2022


Blaize and Innovatrics Provide Edge Ready, Low Power, Low Latency Facial Recognition Technology

Innovatrics and Blaize announced a technology partnership to deliver ready-to-use facial recognition solutions designed for access control and public security applications leveraging SmartFace Embedded and Blaize edge computing devices. 

Innovatrics SmartFace Embedded can run time-critical operations such as face detection, facial landmarks, and face template extraction through on-edge or on-chip processing. Users can choose and combine the specific processes they need for their intended application while maintaining a small footprint, minimal compute capacity, and scalability. 

Data processed at the edge of the network enables organizations to minimize the need for massive network resources and data transfer to a distant data center. Innovatrics’ proprietary facial recognition algorithm enables a camera to pre-process a video stream and perform face matching through edge-to-cloud computing. Combined with real-time AI computing on the edge Blaize Pathfinder P1600 Embedded System on Module (SoM), SmartFace Embedded can detect facial landmarks within 5ms, face template extraction within 15ms, and face detection within 20ms.   

The availability of SmatFace Embedded facial recognition technology on Blaize devices is a boon to the current environment where a global shortage of GPUs exists. With its edge-to-cloud approach, each camera can be made smart in any given surveillance system, and one edge device can process several video streams. Since host processors are not needed with the Blaize Pathfinder P1600 SoM, users can plug them in and instantly use them for their specific use case.

Catch both the Innovatrics and Blaize teams at the Embedded Vision Summit(Blaize Booth #506) in Santa Clara, California, on May 17-18 and the Embedded World (Innovatrics booth #338 in hall 4) in Nuremberg, Germany, on June 21-23. 

For more information, visit: and



Tiera Oliver, Associate Editor for Embedded Computing Design, is responsible for web content edits, product news, and constructing stories. She also assists with newsletter updates as well as contributing and editing content for ECD podcasts and the ECD YouTube channel. Before working at ECD, Tiera graduated from Northern Arizona University where she received her B.S. in journalism and political science and worked as a news reporter for the university’s student led newspaper, The Lumberjack.

More from Tiera