View on GitHub


MIVisionX toolkit is a set of comprehensive computer vision and machine intelligence libraries, utilities, and applications bundled into a single toolkit. Khronos OpenVX is also delivered with MIVisionX.

YoloV2 using AMD WinML Extension

This application shows how to run tiny yolov2 (20 classes) with MIVisionX RunTime:



Step 1. Get ONNX model

Train your own YoloV2 ONNX model or get it from onnx github. ONNX version 1.3 is recommended.

Step 2. Build the app using MIVisionX_winml_YoloV2.sln on Visual Studio.

Step 3. Run tests

Open up the command line prompt or Windows Powershell and use the following commands to run the tests.


The confidence parameter is an optional parameter which sets the confidence level of detection. Lower the confidence level if the detection is not good enough.


.\MIVisionX_winml_YoloV2.exe –image image\cat.jpg –modelLoc model.onnx

Update parameters

Please update parameters (biases, object names, etc) in /source/Region.cpp, and parameters (dim, blockwd, targetBlockwd, classes, etc) in /source/AnnieYoloDetect.cpp