Join us at the Synopsis ARC Processor Summit. There will be several sessions that discuss Khronos APIs and how they contribute to the Vision and Embedded Systems ecosystems.
Khronos Session: The Vision API Landscape
Time: 5:00 – 5:30 pm Speaker:
Neil Trevett, NVIDIA and Khronos Group President Webpage: Click here
The choice of hardware acceleration APIs for parallel computation and vision processing is complex and rapidly evolving. Many of the industry-standard APIs such as OpenCL and OpenVX have been upgraded, while the industry begins to adopt the new generation of low-level, explicit GPU APIs, such as Vulkan, that tightly integrate graphics and compute. Some of these APIs, like OpenVX and OpenCV, are vision-specific, while others, like OpenCL and Vulkan, are general-purpose. Which one(s) should you use for your project?
Khronos-related Courses, Papers, and Sessions
Advanced Vision Capabilities for Next-Generation SoCs
Time: 11:00am – 12:30 pm Speakers: Bo Wu, Embedded Vision Applications Engineer, Synopsys Webpage: Click here
The availability of specialized high-performance processors is making it possible to integrate vision capabilities into SoCs, giving them the ability to see and interpret their surroundings. These heterogeneous multicore processors support HD resolutions with low power consumption and include specialized vision engines to improve accuracy. Supporting these processors are a range of tools including OpenVX™, OpenCL™, and OpenCV that greatly improve developmental productivity. This presentation details the architecture of an embedded vision processor family and the open source vision tools used to program and ensure efficient resource utilization of the heterogeneous multicore platform.
Using the OpenCL C Kernel Language for Embedded Vision Processors
OpenCL C is a programming language that is used to write computation kernels. This presentation will focus on the benefits and ease of programming vision-based kernels using the key features of OpenCL C. The language extensions that allow programmers to take advantage of hardware features typical of embedded vision processors, such as wider vector widths, sophisticated accumulator forms of instructions, and scatter/gather capabilities, will be described. Advanced topics, such as whole function vectorization support available in the compiler and the benefits of hardware support for predication in the context of lane-based control flow and OpenCL C will also be covered.