Rick Maule, Senior Director of Project Management at Qualcomm, presents the "Using Heterogeneous Computing for Mobile and Embedded Vision" tutorial within the "Implementing Vision Systems" technical session at the October 2013 Embedded Vision Summit East.
A single vision application typically incorporates multiple algorithms requiring many different types of computation. This diversity makes it difficult for a single type of processing engine, such as the CPU, to execute vision applications with maximum efficiency. In this presentation, Qualcomm explores the potential for implementing vision applications on heterogeneous processor chips — that is, chips consisting of multiple processing engines of different architecture types, such as the CPU, GPU, and DSP . Using mobile devices as an example, we investigate the opportunities, challenges, trade-offs and techniques associated with mapping vision applications onto heterogeneous processing engines, particularly for applications requiring maximum energy and thermal efficiency.