The mechanism used by OpenVX framework to add new vision functions to OpenVX by 3rd party vendors

Hi,
I start to learn the OpenVX programming, and then am curious about how to make a newly designed embedded processor to support CV-related application developers via OpenVX. Now, I don’t know the procedure of making the code written in C or intrinsic functions of a new processor to be called by OpenVX APIs. I am grateful for getting answers or suggestions.

@xwentian here is a sample on how to extend OpenVX functionality with vendor extensions.

This extension implements a method to use OpenCV functions within a OpenVX kernel.

more extensions here - https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX/tree/master/amd_openvx_extensions

Hi, Kiriti

Thank you very much for the provided information.

1 Like