Memory Allocation in AI and Computer Vision Applications


Date:  Tuesday, September 22, 2020

Start Time: 12:30 pm

End Time: 1:00 pm

Efficient edge computing is the foundation underlying AI and computer vision applications in autonomous vehicles, cameras, drones and many other applications. Because deep neural networks (DNNs) are memory intensive, creating efficient implementations of DNNs requires efficient use of memory and memory bandwidth. In this talk, we’ll present CEVA’s novel approach to enable efficient memory allocation to enable implementing DNNs under strict size and power constraints. Our approach utilizes a unified computational graph and takes into account the differing characteristics of different classes of memory (on-chip L1 and L2 SRAM and external DDR). We also introduce CEVA’s XM6 and SensPro processors for vision and AI.

There will be a live Q&A session with the presenter immediately following the presentation.

Track

Session Speakers

  • Rami Drucker has served as CEVA’s Neural Graph Compute Lead in the Vision Business Unit since January 2020. Prior to this, Mr. Drucker was a senior Software Architect at OSR Enterprises AG, where he played a key role in the design of EVOLVER, a Multi-Domain AI Brain for next-generation autonomous and securely connected vehicles. Mr. Drucker holds a B.Sc. in Mathematics and Computer Science from Tel-Aviv University and a M.Sc. degree in Computer Science from Bar Ilan University in Israel


Share

Memory Allocation in AI and Computer Vision Applications