Skip to content
This repository has been archived by the owner on Oct 31, 2023. It is now read-only.

DensePose for JS #272

Open
rs9899 opened this issue Jul 1, 2020 · 2 comments
Open

DensePose for JS #272

rs9899 opened this issue Jul 1, 2020 · 2 comments

Comments

@rs9899
Copy link

rs9899 commented Jul 1, 2020

I am using DensePose to get IUV_map which I am planning to use later on the pipeline. The issue is that, for deployment, I need a JS version to run on web. Tensorflow.js does run in browser but DensePose is in pytorch.

Any idea for running the task on browser will be helpful.

Thanks

@aryanbdps9
Copy link

This link suggests that saving the model in ONNX and loading it using Tensorflow is one way to do it. Have you tried it?

@rs9899
Copy link
Author

rs9899 commented Jul 1, 2020

For densepose, documentation for export to ONNX mentions that some of the operations are only available in caffe2 and that will disrupt the conversion further to tensorflow. That's why i wanted feedback if someone found a way to do the same.

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

No branches or pull requests

2 participants