Virtualization Technology News and Information
Article
RSS
VeriSilicon and NXP Collaborate on Machine Learning Across Wide Range of ML-Enabled Devices

Today at the 2019 Consumer Electronics Show VeriSilicon announces the extension of its collaboration with NXP Semiconductors N.V. to machine learning. NXP's upcoming i.MX products will include VeriSilicon's Vivante intelligent pixel processing IP solution centered around the VIP Neural Network Processor Unit.

VeriSilicon Vivante VIP NPU features full scalability from extremely low power sub TOPS to high performance exceeding 100 TOPS. Vivante VIP8000/VIPNano NPU IP has been powering next-generation Intelligent devices across multiple market segments such as smart home, industrial, mobile, and automotive ADAS.

Vivante VIP processor supports the OpenCL 1.2 FP and OpenVX 1.2 APIs and common AI frameworks with a full software suite, including SDKs. All Vivante VIP configurations share a common unified software stack that has been deployed in products in multiple market segments.

"VeriSilicon's Vivante VIP IP family NPU's architecture has full scalability to address AI Everywhere, from battery operated OS-Less, DDR-Less devices to high performance Neural Network processors in Edge Server applications with a common SDK and unified software. We are honored to have NXP as a technology partner to address key market segments with Machine Learning technology. Vivante VIP IP is a leader in the NPU IP market with multiple products in the marketplace and solutions for each of the key market segments," said Weijin Dai, executive vice president and general manager of the Intellectual Property Division at VeriSilicon. "Continuing with our long-term technology partnerships with NXP, this engagement adopted the full Vivante Intelligent Pixel Processing Solution, which includes NPU, ISP, GPU, and Video IP."

"NXP's vision is machine learning everywhere. VeriSilicon Vivante VIP architecture matches our vision," said Martyn Humphries , vice president and general manager of i.MX applications processors at NXP. "These production-proven machine learning IP's together with our machine learning eIQ tools enable us to rapidly develop and deploy cloud-to-edge solutions."

Published Tuesday, January 08, 2019 3:52 PM by David Marshall
Filed under:
Comments
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
Calendar
<January 2019>
SuMoTuWeThFrSa
303112345
6789101112
13141516171819
20212223242526
272829303112
3456789