arvindchandel

I want to try, but issue is that i have to integrate it with deepstream, for that i am looking for a way.

@c7934597 trt_pose with deepstream is not good combination, i tried that, better you try something else.

Hi @c7934597 yes i tried same and re-implemented it in python, but was not satisfied with pose quality so i am trying other state of art, but u can try...

@chaobeiying I am getting same error while converting SlowFast model to onnx. did you get any solution for this?

> There's a lot of questions online about this, for example, the following link might help you, to sum up, you just need to save the TensorFlow graph to disks...

Got it working.

I also faced the same issue and this solution by @Snehal-Reddy works fine.

Hi did u get any solution of the Id switch issue?

I also tried through unzip command and there was error in linux, so unziped in windows, and it was successful. It was also issue for me so sharing my hack....