Companies collaborate to enable CEVA-XM family with Halide functionality, dramatically improving productivity and performance of vision software
MOUNTAIN VIEW, Calif. -- Feb. 23, 2018 -- CEVA, Inc. (NASDAQ: CEVA), the leading licensor of signal processing platforms and artificial intelligence processors for smarter, connected devices, today announced that it has partnered with mPerpetuo Inc., a Bay Area company specializing in the design of imaging devices and their underlying technologies, to enable Halide functionality on the CEVA-XM family of imaging and vision processors.
Halide is a domain-specific language for imaging and vision applications developed by MIT CSAIL and adopted by leading technology companies for use in their imaging products. This language provides abstractions that allow for rapid development of highly optimized and portable imaging applications.
The CEVA-XM family of imaging and vision processors and platforms have achieved more than 50 design wins to date, with many of the world's leading OEMs. The CEVA-XM processors power computer vision and computational photography in millions of devices, including smartphones, drones, mirrorless cameras, 360 degree cameras, action cameras surveillance cameras and VR headsets, and are soon to be deployed in automotive image sensors and ADAS pre-processors.
"Enabling Halide on the CEVA-XM family offers customers a rapid and powerful method of expressing, prototyping, and optimizing complex vision and imaging processing pipelines," said Gary Gitelson, VP of Engineering at mPerpetuo. "When using the Halide development environment, programmers can create software with performance that is on-par or exceeds hand-coded assembly or intrinsics, while achieving significantly reduced development times, and improved code consistency and maintainability."
"Support for Halide will offer significant performance benefits to our customers wishing to write high-performance image processing code for their CEVA-XM powered products," said Ilan Yona, vice president and general manager of the vision business unit at CEVA. "We are pleased to collaborate with mPerpetuo to add support for this exciting new programming language to our industry-leading imaging and vision processors."
mPerpetuo provides a full Halide language port that can generate CEVA-XM assembly code and a full runtime environment to execute those pipelines on the CEVA-XM hardware. These tools and technologies form the core necessary to develop imaging algorithms targeted at the CEVA-XM using the Halide language. In addition, CEVA and mPerpetuo provide a reference imaging pipeline that can be used as a basis for building custom camera acquisition, display, image processing and vision applications. This pipeline contains all of the key basic blocks of image processing from raw sensor input to denoised and color processed final output. It is highly customizable and built in a modular fashion in which new blocks can be added easily, and existing blocks can be modified or replaced as needed according to a customer's design. On top of this baseline image pipeline, mPerpetuo also provides additional visual code generation tools to ease the process of software development with Halide, and consultation in optimizing and working with Halide on the CEVA-XM family of products. mPerpetuo and CEVA will join forces to train and enable their joint customers to migrate their ISPs to Halide.
mPerpetuo Inc. (www.mperpetuo.com) is a Bay Area start-up, that has set out to solve real-world system challenges of designing and optimizing image capture and display devices using vision, image processing, and real-time technologies. For more information and or live demonstrations of mPerpetuo's portfolio, please contact at firstname.lastname@example.org.
About CEVA, Inc.
CEVA is the leading licensor of signal processing platforms and artificial intelligence processors for a smarter, connected world. We partner with semiconductor companies and OEMs worldwide to create power-efficient, intelligent and connected devices for a range of end markets, including mobile, consumer, automotive, industrial and IoT. Our ultra-low-power IPs for vision, audio, communications and connectivity include comprehensive DSP-based platforms for LTE/LTE-A/5G baseband processing in handsets, infrastructure and machine-to-machine devices, advanced imaging and computer vision for any camera-enabled device, audio/voice/speech and ultra-low power always-on/sensing applications for multiple IoT markets. For artificial intelligence, we offer a family of AI processors capable of handling the complete gamut of neural network workloads, on-device. For connectivity, we offer the industry's most widely adopted IPs for Bluetooth (low energy and dual mode) and Wi-Fi (802.11 a/b/g/n/ac/ax up to 4x4). Visit us at www.ceva-dsp.com