CEVA Introduces Deep Neural Network Framework to Accelerate Machine Learning Deployment in Low-Power Embedded Systems

SANTA CLARA, Calif -  Linley Processor Conference 2015 -- CEVA, Inc. (NASDAQ: CEVA), the leading licensor of DSP IP platforms for cellular, multimedia and connectivity, has today introduced the CEVA Deep Neural Network (CDNN), a real-time neural network software framework, to streamline machine learning deployment in low-power embedded systems. Harnessing the processing power of the  CEVA-XM4 imaging & vision DSP, the CDNN enables embedded systems to perform deep learning tasks 3x faster than the leading GPU-based systems while consuming 30x less power and requiring 15x less memory bandwidth*. For example, running a Deep Neural Network (DNN) based pedestrian detection algorithm at 28nm requires less than 30mW for a 1080p 30 frames per second video stream.

Key to the performance, low power and low memory bandwidth capabilities of CDNN is the CEVA Network Generator, a proprietary automated technology that converts a customer's network structure and weights to a slim, customized network model used in real-time. This enables a faster network model which consumes significantly lower power and memory bandwidth, with less than 1% degradation in accuracy compared to the original network. Once the customized embedded-ready network is generated, it runs on the CEVA-XM4 imaging and vision DSP using fully optimized Convolutional Neural Network (CNN) layers, software libraries and APIs.

Phi Algorithm Solutions, a member of CEVA's  CEVAnet partner program, has used CDNN to implement a CNN-based Universal Object Detector algorithm for the CEVA-XM4 DSP. This is now available for application developers and OEMs to run a variety of applications including pedestrian detection and face detection for security, ADAS and other embedded devices based around low-power camera-enabled systems.

"The CEVA Deep Neural Network framework provided a quick and smooth path from offline training to real-time detection for our convolutional neural network based algorithms," said Steven Hanna, president and co-founder at Phi Algorithm Solutions. "In a matter of days we were able to get an optimized implementation of our unique object detection network, while significantly reducing power consumption compared to other platforms. The CEVA-XM4 imaging & vision DSP together with the CDNN framework is ideal for embedded vision devices and paves the way to advances in artificial intelligence devices in the coming years using deep learning techniques."

"With more than 20 design wins to-date, we continue to lead the industry in the embedded vision processor domain and are constantly enhancing our portfolio of vision IP offerings to help our customers get to market quicker with minimal risk," said Eran Briman, vice president of marketing at CEVA. "Our new Deep Neural Network framework for the CEVA-XM4 is the first of its kind in the embedded industry, providing a significant step forward for developers looking to implement viable deep learning algorithms within power-constrained embedded systems."

The CDNN software framework is supplied as source code, extending the CEVA-XM4's existing  Application Developer Kit (ADK). It is flexible and modular, capable of supporting either the complete CNN implementation or specific layers. It works with various networks and structures, such as networks developed with  CaffeTorch or  Theano training frameworks, or proprietary networks. CDNN includes real-time example models for image classification, localization and object recognition. It is intended to be used for object and scene recognition, advanced driver assistance systems (ADAS), Artificial intelligence (AI), video analytics, augmented reality (AR), virtual reality (VR) and similar computer vision applications. For more information on CDNN, go to  http://launch.ceva-dsp.com/cdnn/.

On November 12th, CEVA will host a live webinar on implementing machine vision in embedded systems, including a deep dive into CDNN. For more information and to register for the webinar, visit  http://bit.ly/1hoam9p.

CEVA will also present CDNN at the Linley Processor Conference 2015, taking place today in Santa Clara, California. For more information, visit the Linley Processor Conference website at:  http://www.linleygroup.com/events/event.php?num=35.

*Comparison run on AlexNet network, the most popular deep neural network. 

About CEVA, Inc.

CEVA is the leading licensor of cellular, multimedia and connectivity technologies to semiconductor companies and OEMs serving the mobile, consumer, automotive and IoT markets. Our DSP IP portfolio includes comprehensive platforms for multimode 2G/3G/LTE/LTE-A baseband processing in terminals and infrastructure, computer vision and computational photography for any camera-enabled device, audio/voice/speech and ultra-low power always-on/sensing applications for multiple IoT markets. For connectivity, we offer the industry's most widely adopted IPs for Bluetooth (Smart and Smart Ready), Wi-Fi (802.11 b/g/n/ac up to 4x4) and serial storage (SATA and SAS). One in every three phones sold worldwide is powered by CEVA, from many of the world's leading OEMs including Samsung, Huawei, Xiaomi, Lenovo, HTC, LG, Coolpad, ZTE, Micromax and Meizu. Visit us at  www.ceva-dsp.com and follow us on  TwitterYouTube and LinkedIn.




Review Article Be the first to review this article
Featured Video
Editorial
Peggy AycinenaWhat Would Joe Do?
by Peggy Aycinena
Qualcomm’s Lu Dai: Energetic leadership for Accellera
More Editorial  
Jobs
SOC Logic Design Engineer for Global Foundaries at Santa Clara, CA
Technical Support Engineer for EDA Careers at Freemont, CA
Principal Engineer FPGA Design for Intevac at Santa Clara, CA
Development Engineer-WEB SKILLS +++ for EDA Careers at North Valley, CA
Sr. Staff Design SSD ASIC Engineer for Toshiba America Electronic Components. Inc. at San Jose, CA
Upcoming Events
IoT Summit 2017 at Great America ballroom, Santa Clara Convention Center Santa Clara CA - Mar 16 - 17, 2017
SNUG Silicon Valley 2017 at Santa Clara Convention Center Santa Clara CA - Mar 22 - 23, 2017
CDNLive Silicon Valley 2017 at Santa Clara Convention Center Santa Clara CA - Apr 11 - 12, 2017
10th Anniversary of Cyber-Physical Systems Week at Pittsburgh, PA, USA PA - Apr 18 - 21, 2017



Internet Business Systems © 2017 Internet Business Systems, Inc.
595 Millich Dr., Suite 216, Campbell, CA 95008
+1 (408)-337-6870 — Contact Us, or visit our other sites:
AECCafe - Architectural Design and Engineering TechJobsCafe - Technical Jobs and Resumes GISCafe - Geographical Information Services  MCADCafe - Mechanical Design and Engineering ShareCG - Share Computer Graphic (CG) Animation, 3D Art and 3D Models
  Privacy Policy