CEVA-XM6 and CEVA Deep Neural Network software framework (CDNN) enables advanced computer vision and AI at the edge in Sigmastar’s SAV538 AI Camera SoC targeting surveillance, automotive, and virtual reality applications

CEVA Technology Symposium – CEVA, Inc. (NASDAQ: CEVA), the leading licensor of signal processing platforms and artificial intelligence processors for smarter, connected devices, today announced that SigmaStar Technology Corp., a wholly-owned subsidiary of MStar Semiconductor Inc., has licensed and deployed the CEVA-XM6 computer vision and deep learning platform to enable advanced computer vision and neural-network based applications in its SAV538 Artificial Intelligence (AI) camera System-on-Chip (SoC).

To commemorate the strong partnership between the two companies, Robert Lin, CEO of SigmaStar will deliver a keynote speech at the CEVA Technology Symposium Shenzhen event, taking place on October 31. For more information please visit https://events.ceva-dsp.com/symposium-2018-en/.

The SAV538 SoC incorporates the CEVA-XM6 and CEVA Deep Neural Network (CDNN) technologies to enable intelligent video content analytics for a broad range of camera-based applications including surveillance, automotive and virtual reality. The CEVA platform efficiently supports a combination of computer vision algorithms and neural networks, providing SigmaStar customers with a differentiated, programmable solution on which to develop their AI-based camera applications. The SoC is readily available today and is already being adopted by multiple OEMs/ODMs for a range of intelligent camera applications.

“The SAV538 intelligent camera SoC delivers outstanding performance and ultra-low power consumption for anyone wishing to embed vision-based AI in their device,” said HM Cheng, associate vice president at SigmaStar. “CEVA’s computer vision and neural network platform is key to our SoC, allowing us and our customers to simply and efficiently deploy high-performance computer vision algorithms and neural networks on the edge device, without requiring a connection to the cloud.”

“With the adoption of our CEVA-XM6 and CDNN technologies, SigmaStar has delivered on the promise of AI at the edge in their intelligent camera SoC, bringing advanced video content analytics to the mass market,” said Ilan Yona, vice president and general manager of CEVA’s Vision Business Unit. “We’re proud to have collaborated closely with SigmaStar in the development of this outstanding product and look forward to continuing our relationship with them as AI becomes a ‘must-have’ technology in every camera-enabled device.”

CEVA’s latest generation imaging and vision platforms address the extreme processing requirements and low power constraints of the most sophisticated machine learning and machine vision applications used in smartphones, surveillance, AR & VR, drones and self-driving cars. These DSP-based platforms include a hybrid architecture composed of scalar and vector DSP processors and hardware accelerators, coupled with a comprehensive Application Development Kit (ADK) to streamline software deployment. The CEVA ADK includes: CEVA-Link for seamless software level integration with a host processor; optimized software libraries for widely used imaging and computer vision algorithms; the CEVA Deep Neural Network (CDNN) real-time Neural Network software framework which streamlines machine learning deployment at a fraction of the power consumption of the leading GPU-based systems, and; state-of-the-art development and debugging tools. For more demanding neural network workloads, the CEVA-XM processors can be paired with a NeuPro™ processor, CEVA’s family of specialized AI processors for deep learning at the edge.
For more information, visit http://www.ceva-dsp.com/app/imaging-computer-vision/ or https://www.ceva-dsp.com/product/ceva-neupro/