model post-processing modify

[sunbin1357]

I want to use your provided c++ code for model post-processing, but my keypoint is 19 and paf is 38, What parts of your provided c++ code do I need to modify.

[jaybdub]

Hi sunbin1357,

Thanks for reaching out!

You should not need to modify the functions directly, but just provide different arguments to match your configuration.

There are some more notes here

https://github.com/NVIDIA-AI-IOT/trt_pose/tree/master/trt_pose/parse

I'd get it to work in Python first, using the bindings for these functions.

You can see how these methods are called here

https://github.com/NVIDIA-AI-IOT/trt_pose/blob/master/trt_pose/parse_objects.py#L17

The toplology can be generated from the COCO annotation

https://github.com/NVIDIA-AI-IOT/trt_pose/blob/master/trt_pose/coco.py#L21

Please let me know if this helps or you run into any issues.

Best, John

[sunbin1357]

According to your suggestion, I modified the corresponding code, but the result is shown in the below

[thancaocuong]

@sunbin1357 can you show me your modifications? I will check it out and give you some suggestions

[sunbin1357]

I add tensor.contiguous(), and the problem is solved

thancaocuong [email protected] 于2020年8月10日周一 下午12:53写道：

@sunbin1357 https://github.com/sunbin1357 can you show me your modifications? I will check it out and give you some suggestions

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/NVIDIA-AI-IOT/trt_pose/issues/58#issuecomment-671160924, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALNJVR3IUDOLDCCHRJRG223R754MNANCNFSM4PKHI5OA .