The Raspberry Pi Foundation has announced it's bringing the OpenVX 1.3 API to Raspberry Pi devices to improve computer vision on the popular single-board computers.
The new open and royalty-free API comes from the Khronos Group, which has backed standards like Vulcan and OpenCL. Khronos members include most big-name software and hardware vendors – AMD, Apple, Arm, Epic Games, Google, Samsung, Intel, Nvidia and so on – as well as companies with a stake in its standards, like Boeing and IKEA.
Developer
"The Khronos Group and Raspberry Pi have come together to work on an open-source implementation of OpenVX 1.3, which passes the conformance on Raspberry Pi," explained Kiriti Nagesh Gowda, AMD's MTS software development engineer.
SEE: Virtual hiring tips for job seekers and recruiters (free PDF) (TechRepublic)
"The open-source implementation passes the Vision, Enhanced Vision, & Neural Net conformance profiles specified in OpenVX 1.3 on Raspberry Pi."
The Raspberry Pi, being popular, cheap and relatively low powered, is an important device to prove conformance for the computer-vision standard.
Gowda explains that the OpenVX 1.3 conformance testing conducted for Raspberry Pi "helps ensure that Khronos standards are consistently implemented by multiple vendors to create a reliable platform for developers".
OpenVX aims to bring efficient, real-time computer-vision processing to embedded computer devices, such as surveillance devices, advanced driver assistance systems, augmented reality, and robotics.
"The developers can take advantage of using this robust API in their application and know that the application is portable across all the conformant hardware," explained Gowda.
The two Raspberry Pi models that passed the test are the Raspberry Pi 3 Model and the Raspberry Pi 4 Model B.
The Khronos Group has published a OpenVX 1.3 sample implementation as well as several sample computer-vision applications on GitHub.
The OpenVX 1.3 implementation is available on GitHub.