Dear MediaTek community,
The ARM Mali GPUs on Genio support OpenCL. I am interested to find out if parallel computing can be achieved using OpenCL in a similar manner as Nvidia CUDA on Nvidia GPUs. And if there is any documentation available for this.
The motivation behind parallel computing is using OpenCV, OpenCL to delegate Tasks to GPU, CPU and where applicable to NPU…
OpenCV already comes with Support for OpenCL.
Hello @Salman_s,
That is a great idea!
We have successfully used OpenCL accelerated OpenCV on a Genio 510 project from one of our clients and it lowered CPU usage dramatically for the custom application.
In order to use OpenCL acceleration, you will need to add OpenCL support to the OpenCV recipe.
In order to achieve this, we encourage you to follow our short wiki:
How to enable OpenCL support in OpenCV for MediaTek Yocto platforms
Please let us know if you have other questions or require further support, we would love to help.
best regards.
Andres
Embedded Software Engineer at ProventusNova