Open standard for portable, performance and power-optimized vision applications and libraries; Provisional Specification for Public Review
November 19th 2013 – SIGGRAPH Asia – Hong Kong – The Khronos Group today announced the ratification and public release of the OpenVX 1.0 provisional specification, an open, royalty-free standard 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. The provisional release of the specification enables developers and implementers to provide feedback before specification finalization, which is expected within six months. The OpenVX 1.0 provisional specification is available at www.khronos.org/openvx.
“Computer vision is the biggest, most disruptive, application segment in technology today. From automotive, to security, to consumer capture with 3D sensors, 4K sensors, and sensors so small they can be put in UAVs the size of a fly, the processing of photons has never been more challenging,” said Dr. Jon Peddie, president of Jon Peddie Research. “Being able to manage, process, and quickly move sensor data without consuming much power is critical and only OpenVX offers the mechanisms necessary to balance all those issues - it’s going to change the way we do vision systems.”
OpenVX enables significant implementation innovation while maintaining a consistent API for developers. An OpenVX application expresses vision processing holistically as a graph of function nodes. An OpenVX implementer can optimize graph execution through a wide variety of techniques such as: acceleration of nodes on CPUs, GPUs, DSPs or dedicated hardware, compiler optimizations, node coalescing, and tiled execution to keep sections of processed images in local memories as they flow through the graph. Khronos has released a provisional tiled execution extension alongside the main OpenVX specification to enable user custom kernels to exploit this style of optimization. Additionally, Khronos has released the VXU™ utility library to enable developers using OpenVX to call individual nodes as standalone functions for easy code migration.
“Computer vision is central to bringing natural user interfaces and environmental awareness to mobile devices and OpenVX enables cross-platform processing with the high performance and low power that will be vital to widespread adoption,” said Neil Trevett, president of the Khronos Group and vice president of mobile content at NVIDIA. “OpenVX has been designed to be implemented independently or to interoperate within the Khronos standards ecosystem for camera control, sensor fusion, data flow, compute acceleration and graphics rendering, ensuring Khronos APIs continue to meet the needs of portable, state-of the art applications.”
OpenVX can be used directly by applications or to accelerate higher-level middleware, such as the popular OpenCV open source vision library that is often used for application prototyping. OpenVX will have extensive conformance tests to complement a focused and tightly defined finalized specification for consistent and reliable operation across multiple vendors and platforms making OpenVX an ideal foundation for shipping production vision applications. Finally, as any Khronos specification, OpenVX is extensible to enable nodes to be deployed to meet customer needs, ahead of being integrated into the core specification.
“The release of OpenVX 1.0 is a ground-breaking step that will accelerate mass market adoption of computer vision applications in mobile, home, automotive, and embedded products. The specification benefits developers and lays a foundation for complex vision algorithms to be simplified and power/performance/bandwidth optimized on OpenVX compliant hardware, enabling novel uses of vision processing on any platform. Participation in the workgroup has allowed us to make significant breakthroughs in our Vega GPUs to streamline the visual processing pipeline from beginning to end,” said Wei-Jin Dai, Vivante CEO.
“The Itseez team is excited about the release of the OpenVX 1.0 provisional specification. It will enable speed- and power-optimized implementations of OpenCV across a wide range of mobile and embedded platforms, stimulating growth for the computer vision industry and the development of new great applications,” said Victor Erukhimov, CEO, Itseez and chair of the OpenVX working group.
“CEVA extends its congratulations to the Khronos Group on the release of the OpenVX 1.0 specification, which sets the foundation for mass market adoption of computer vision applications across multiple industries,” said Eran Briman, vice president of marketing at CEVA. “In particular, OpenVX directly addresses the power consumption challenges faced when implementing complex vision algorithms in power-sensitive products by enabling the seamless offload of these tasks from the CPU and GPU onto dedicated vision engines, such as our CEVA-MM3101 platform, resulting in significant power savings.”
“Going forward, vision systems will be a key differentiator for a wide range of consumer products including smartphones, tablets, automotive driver assistance and many more. Based on Imagination’s in-depth imaging expertise, we are offering innovative PowerVR products for imaging and vision that will enable our customers to integrate this functionality on the SoC We are pleased to see Khronos taking a leading role in driving open standards for computer vision. OpenVX 1.0 is an important starting point for accelerating creation and adoption of a wide range of vision applications,” said Peter McGuinness, director of multimedia technology marketing, Imagination Technologies.
“Movidius anticipates OpenVX will stimulate incredible innovation as it enables cross-platform, scalable computer vision for mobile devices,” said Remi El-Ouazzane, CEO of Movidius. “In combination with OpenVX capable applications, Movidius’ computational imaging chipsets empower mobile developers to deploy vision-based applications that were simply never possible before.”
“videantis congratulates the Khronos Group on reaching this major milestone. OpenVX enables efficient acceleration of computer vision algorithms, a key technology driving new applications such as always-on camera applications, gesture interfaces, and automotive driver assistance systems. We’re proud to bring support for this new standard to our v-MP4000HDX scalable unified video/vision processor architecture,” said Hans-Joachim Stolberg, CEO at videantis.