search

Samleo8 / learnable-triangulation-pytorch

The repository that this is forked from is an official PyTorch implementation of the paper "Learnable Triangulation of Human Pose" (ICCV 2019, oral). Proposed method archives state-of-the-art results in multi-view 3D human pose estimation! This repo aims to extend the capabilites to the CMU dataset
MIT License
27 stars 11 forks source link

Visualise results #3

Open agenthong opened 3 years ago

agenthong commented 3 years ago

It's a great repo. In your code there is a 'visualise_results' file, I run it but I only see this: image

Is it correct and can I see the 3D results?

Thanks!

Samleo8 commented 3 years ago

Hi sorry, it has been a whlie since I have worked on the repository. If I recall correctly, the visualise_results.py file does not visualise things in 3D, sorry :(

It only takes the results in your results.pkl from running the network and displays it in the form you see above (i.e. from different angles). It was an earlier attempt to get over issues with Tensorboard, but the same results can be seen in tensorboard.

I (or you!) could attempt to update the visualiser using matplotlib.pyplot 3d plots, but I have since lost touch with the repo, and it'll take a while for me to refamiliarise.

anas-zafar commented 1 year ago

Hi @agenthong , can you please guide me how did you run the above visualization? Thanks