How can I use my TensorRT engine with a video file with the trtexec command ?

I converted my yolo model to onnx using yolov3_to_onnx.py. than I converted yolov3.onnx to yolov3.engine for TensorRT C++ API. I want to test yolo.engine with video file. I tried to using trtexec command but I couldn’t. How can I test on video file using with my yolo engine ?

Hi,

I think “trtexec” doesn’t support a video file as input.

Please refer to below sample example:
https://github.com/NVIDIA/object-detection-tensorrt-example

Thanks

Hi,

Do you have anything yolo-tensorRT example for tensorRT C++ API ?

thanks for your help.

Hi,

Currently we just have yolo-tensort sample using python API.
https://docs.nvidia.com/deeplearning/sdk/tensorrt-archived/tensorrt-700/tensorrt-sample-support-guide/index.html#yolov3_onnx

But after converting yolo to onnx you can refer following sample for ONNX to TRT conversion:
https://github.com/NVIDIA/TensorRT/tree/release/7.0/samples/opensource/sampleOnnxMNIST

Thanks

hi,

Do you have anything tensorRT C++ API example for video file ?

thanks

Hi,

Please refer to below code samples, hope it helps:
https://github.com/dusty-nv/jetson-inference#code-examples

Thanks