Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

posts/pytorch-train-keypoint-rcnn-tutorial/index #53

Open
utterances-bot opened this issue Sep 18, 2024 · 3 comments
Open

posts/pytorch-train-keypoint-rcnn-tutorial/index #53

utterances-bot opened this issue Sep 18, 2024 · 3 comments

Comments

@utterances-bot
Copy link

Christian Mills - Training Keypoint R-CNN Models with PyTorch

Learn how to train Keypoint R-CNN models on custom datasets with PyTorch.

https://christianjmills.com/posts/pytorch-train-keypoint-rcnn-tutorial/index.html

Copy link

Hi Christian,
thanks for your greate work. Could you also suggest, how to speed up the inference time? You already export the model to onnx but let us also know how to convert to TensorRT and also other ways to improve speed.

@cj-mills
Copy link
Owner

Hi @Masih-Zoharian,

Based on my personal experience and some NVIDIA forum threads, I don't believe the Keypoint R-CNN model included with torchvision is currently supported by TensorRT.

From my experience even ONNX Runtime's tools for collecting the calibration data used for quantizing the model don't seem set up to support the Keypoint R-CNN model.

I recommend trying a different keypoint detection model if you need something more efficient.

@Masih-Zoharian
Copy link

Masih-Zoharian commented Sep 20, 2024 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants