OpenVX is a state-of-the-art open API standard for accelerating applications using computer vision and machine learning. The API and its conformance tests enable applications to leverage highly specialized features of hardware platforms while still retaining portability of application code across a wide range of architectures. This talk will use concrete examples on real implementations to demonstrate the performance portability of OpenVX. Example applications written using OpenVX will be described that run on platforms developed by Cadence Design Systems, Texas Instruments, Advanced Micro Devices and Axis Communications. Benchmarks will demonstrate performance gains that would otherwise only be achievable via hardware-specific code optimizations. The talk will also provide an update on the new features of the latest version of the OpenVX API, including support for a cross-platform neural network inferencing engine standard using a combination of OpenVX and Khronos’ Neural Network Exchange Format (NNEF).
Portable Performance via the OpenVX Computer Vision Library: Case Studies
Session Speakers
-
Frank Brill
Design Engineering Director, Cadence
Frank Brill manages OpenVX software development for Cadence’s Tensilica Imaging and Vision DSP organization. He began his career doing computer vision research and development for video surveillance applications at Texas Instruments, and then moved into silicon device program management, where he was responsible for several digital still camera and multimedia chips. Since then, Frank has managed computer vision R&D groups at TI, NVIDIA, Samsung and now at Cadence, and has represented all four companies in the Khronos OpenVX working group. He joined Cadence in 2016 to work on OpenVX, and currently serves as chairperson of the OpenVX working group.
