CVPR - OpenVX Tutorial

Event is now over

Checkout some of our upcoming events over here.

June 7, 2015 Afternoon
Hynes Convention Center, Room 104, Boston, MA

CVPR is the premier annual Computer Vision event comprising the main CVPR conference and several co-located workshops and short courses. Come to the session on OpenVX!

OpenVX: a framework for accelerating computer vision
CVPR 2015 tutorial

OpenVX is an open standard released by the Khronos Group in 2014. OpenVX enables performance and power-optimized computer vision functionality processing, especially important in embedded and real-time use cases. We will cover graph API that enables OpenVX developers to efficiently run computer vision algorithms on heterogeneous architectures. A set of example algorithms from computational photography and advanced driver assistance mapped to the graph API will be discussed. We will cover the relationship between OpenVX and OpenCV, as well as OpenCL. The second half of the tutorial will be a practice session, dedicated to solving a computer vision problem with OpenVX using a conformant OpenVX implementation.

Practice session
We will work on a tracking algorithm implemented with OpenVX. In the end of the session, participants will know how to develop code with OpenVX, and optimize for performance. Please see the requirements for the practice session and please sign up using the form below!

Organizers

Victor Erukhimov, Itseez
Victor is a co-founder of Itseez, Inc. and has an extensive background in computer vision. Prior to joining Itseez, Victor worked as a project manager and senior research scientist at Intel Corporation. He is the author of more than 25 papers in the areas of computer vision and machine learning as well as several US and international patents. Victor has also been involved in several open source projects, being a developer of OpenCV library. Since 2012, Victor has served as chair of the OpenVX working group at Khronos that develops the open standard for the computer vision industry.

Kari Pulli, Light
Kari is VP of Computational Imaging at Light. Earlier, he was Sr. Director of Research at NVIDIA, and before that Nokia Fellow at Nokia Research center; in both places he headed a research team called Mobile Visual Computing. Kari has a long background in standardization, at Khronos he has contributed to many mobile media standards, including OpenVX. He is a frequent author and speaker at venues like CVPR and SIGGRAPH, with h-index of 25. Kari has a PhD from University of Washington, MBA from University of Oulu, and has taught and worked as a researcher at University of Oulu, Stanford University, and MIT.

Thierry Lepley, NVIDIA
Thierry Lepley is Senior Computer Vision Engineer at NVIDIA and the NVIDIA representative in the OpenVX standardization group. His focus is on the development of optimized computer vision toolkits and libraries for real-time embedded systems. Earlier, Thierry was Principal Engineer at STMicroelectronics, working on many-core acceleration for computer vision, where he developed a compiler that automates the parallelization of image processing pipelines and the management of image tiling.

Radhakrishna Giduthuri, AMD
Radhakrishna Giduthuri is a PMTS Design Engineer at AMD focusing on development of computer vision toolkit and libraries for heterogeneous compute platforms. He has extensive background with software design and performance tuning for various computer architectures ranging from General Purpose DSPs, Customizable DSPs, Media Processors, Heterogeneous Processors, GPUs, and several CPUs. He is a member of OpenVX working group representing AMD. In the past he was a member of SMPTE video compression standardizing committee for several years. He is also chair of IEEE Signal Processing Society Chapter of Santa Clara Valley.

Register online

Image Copyright xynntii. Attribution-NonCommercial-ShareAlike 2.0 Generic



Conference Code of Conduct: The Khronos Group is dedicated to providing a harassment-free conference experience for everyone. Visit our Code of Conduct page to learn more.