Invited Presentation: Implementing Image Pyramids Efficiently in Software

Wednesday, May 23, 1:30 PM - 2:00 PM
Summit Track: 
Technical Insights II
Room 203/204

An image pyramid is a series of images, derived from a single original image, wherein each successive image is at a lower resolution than its predecessors. Image pyramids are widely used in computer vision, for example to enable detection of features at different scales. After a brief introduction to image pyramids and their uses in vision applications, we’ll explore techniques for efficiently implementing image pyramids on various processor architectures, including CPUs and GPUs. We’ll illustrate the use of fixed- and floating-point arithmetic, vectorization and parallelization to speed up image pyramid implementations. We’ll also examine how memory caching approaches impact the performance of image pyramid code and discuss considerations for applications requiring real-time response or minimum latency.


Michael Stewart

Proprietor, Polymorphic Technologies

Michael Stewart is a Silicon Valley software professional with extensive imaging and video expertise including highly parallel real-time image processing optimizations; GPU, multi-core SIMD, vectorized DSPs, and custom hardware. His designs have been in ultra-mission critical environments such as hospital cardiac catheterization labs, GI procedure rooms, and on-air broadcast TV stations. He has developed software for computers large and small, from mainframes to SOCs. He has imaging related patents granted and pending.

Complementing his image processing background Michael also has extensive photographic experience and is an exhibiting photographer with shows at several local venues. He has a special interest in imaging algorithms designed to complement and interact with the human physiological perception system.

See you at the Summit! May 20-23 in Santa Clara, California!
Register today and reserve your hotel room!