OpenVX & OpenCL boost performance up to 100X in Synopsys’ Next-Generation Embedded Vision Processors
Synopsys, Inc. announced the DesignWare EV6x family, its newest generation of processor cores optimized for embedded vision applications requiring high definition resolutions. The EV6x processor family is supported by a comprehensive software programming environment based on common embedded vision standards including OpenVX and OpenCL C, as well as Synopsys' MetaWare Development Toolkit. A complete software programming environment including OpenCV and OpenVX libraries, OpenVX runtime, and Synopsys' MetaWare Development Toolkit with OpenCL C vectorizing compiler speeds the development of application software for the DesignWare EV6x processor family. The OpenVX framework simplifies the programming of the EV6x processors by automatically distributing tiled kernel execution over multiple scalar, vector DSP and CNN execution units. The OpenVX framework includes 43 standard computer vision kernels that have been optimized to run on the EV6x Processors. The MetaWare OpenCL C compiler offers automatic whole function vectorization for data-level parallelism and is used for creating kernels for the vector DSPs.
Itseez will become a key ingredient for Intel’s Internet of Things Group (IOTG) roadmap, and will help Intel’s customers create innovative deep-learning-based CV applications like autonomous driving, digital security and surveillance, and industrial inspection. Itseez is also a key contributor to computer vision standards initiatives including OpenCV and OpenVX. Together, we’ll step up our contribution to these standards bodies – defining a technology bridge that helps the industry move more quickly to OpenVX-based products.
Khronos sponsored a day long course covering both the function-based API and the graph API that enable OpenVX developers to efficiently run computer vision algorithms on heterogeneous computing architectures. One section explains the tutorial exercises with a VirtualBox VM, which can be downloaded from the tutorial on Github. The Embedded Vision Summit schedule is located here, the tutorial is on Github and the associated videos from the day long tutorial are available on Youtube.
ELVEES companies group and Imagination Technologies announce the ELVEES Image Semantic Engine (ELISE), a unique next-generation multi-platform SoC offering a wide range of video analytics markets. Companies can use the new ELISE chip to develop a new generation of IP cameras with video analysis and a wide range of OEM modules. Unique hierarchical ELISE tools have been developed, including libraries for the Khronos OpenVX API and the innovative VIPE tool for vision and general purpose algorithm development on mixed multi-platform CPU/Velcore/GPU SoC systems.
The Khronos Group announces the immediate availability of the OpenVX 1.1 specification for cross platform acceleration of computer vision applications and libraries. OpenVX enables performance and power optimized computer vision algorithms for use cases such as face, body and gesture tracking, smart video surveillance, automatic driver assistance systems, object and scene reconstruction, augmented reality, visual inspection, robotics and more. Conformant OpenVX 1.0 implementations and tools are shipping from AMD, Imagination, Intel, NVIDIA, Synopsis and VeriSilicon. OpenVX 1.1 builds on this momentum by adding new processing functions for use cases such as computational photography, and enhances application control over how data is accessed and processed. An open source OpenVX 1.1 sample implementation and full conformance tests will be available in the first half of 2016.
Khronos is proud to be a sponsor of the Embedded Vision Summit on May 2-4 2016 in Santa Clara, CA. The Khronos OpenVX working group will be running an in-depth OpenVX tutorial at this highly focused event. The OpenVX workshop is being held on the final day of the Embedded Vision Summit, and includes tutorials on the OpenVX architecture, a hands-on lab to try out real-world examples of OpenVX acceleration, and tips to deploy OpenVX in your own applications and products.
NVIDIA VisionWorks toolkit is a software development package for computer vision (CV) and image processing. VisionWorks implements and extends the Khronos OpenVX standard, and it is optimized for CUDA-capable GPUs and SOCs enabling developers to realize CV applications on a scalable and flexible platform.
AMD has released a highly optimized open source implementation of the Khronos OpenVX computer vision specification for x86 CPUs and GPUs using OpenCL. The release also includes an open source “graph optimizer” that inspects an OpenVX graph and removes/replaces/merges functions to improve performance and minimize bandwidth. Scripting tools enable rapid prototyping and enables calling any OpenCV functions from within an OpenVX application
Imagination Technologies announces that its PowerVR Rogue Series6 GPUs have achieved Khronos OpenVX 1.0.1 conformance. With OpenVX, developers can run fully optimized vision algorithms on highly parallel, power-efficient PowerVR GPUs that are used in many of the world’s leading smartphones, tablets and other products.
Neil Trevett delivers the presentation, "Update on Khronos Open Standard APIs for Vision Processing," at the December 2015 Embedded Vision Alliance Member Meeting. Neil provides an update on recent developments in multiple Khronos standards useful for vision applications.