YOLO v3 and its tiny version on Xavier

Dear all,

Am searching for a complete guide to run YOLO v3 and its tiny version for custom objects as optimum as possible on Xavier. I appreciate any guidance on this.

Hi,

RidgeRun has a Gstreamer plugin to run inference (Classification and detection) for some networks, TinyYoloV3 is one of the tested networks and it is able to run on CPU or GPU.

For more information please check the following link about GstInference:
https://developer.ridgerun.com/wiki/index.php?title=GstInference/Introduction

Feel free to contact us.

Regards.

Hi,

Have you checked our Deepstream SDK before?
Here is a tutorial for a customized YOLO network with Deepstream SDK:
https://docs.nvidia.com/metropolis/deepstream/4.0/Custom_YOLO_Model_in_the_DeepStream_YOLO_App.pdf

Thanks.