A parallel programming framework for CPU and GPU hardware target platforms
Ben Lever (NICTA)
COMPUTER VISION AND ROBOTICS SERIESDATE: 2011-03-21
TIME: 15:00:00 - 16:00:00
LOCATION: NICTA - 7 London Circuit
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
This talk will provide an introduction to a parallel programming framework being developed as part of the Scalable Vision Machines project. The framework simplifies the task of implementing algorithms that target a range of hardware platforms including multi-core CPUs and GPUs. A number of examples will be shown that describes how computer vision and machine learning algorithms can be specified using this approach.
BIO:
Ben Lever is currently working on the Scalable Vision Machines project at NICTA which has the goal of a developing a parallel programming framework for targeting a variety of hardware platforms. His background is in code-generation for the purpose of high-speed embedded systems simulations, but is now applying some of these techniques for high-performance and highly-parallel architectures.
