Riccardo Marin

RTX 2080 Ti Driver: 460.20 (27.21.14.6020), 17/09/2020

@tomguluson92 , I faced the same issue and solved with these lines in "test_LVD_MANO.py": ``` #d1 = torch.sqrt(directed_distance(vertices_scan_torch, vertices_smpl[0], False)).mean() #d2 = torch.sqrt(directed_distance(vertices_smpl[0], vertices_scan_torch, False)).mean() d = sided_distance(torch.unsqueeze(vertices_scan_torch,0), torch.unsqueeze(vertices_smpl[0],0))[0] #d1...

Hi @boqian-li , Thank you for your interest in our work! This error looks familiar to me- it should have been caused by a wrong segmentation of the mesh using...

I'm happy that it worked! :) The training code is available under the "train" branch of this repository. It is still a bit messy (I plan to polish it in...

You can find some commands to install and run the training here: https://github.com/riccardomarin/NICP/tree/train/training You can run it after the data preprocessing: https://github.com/riccardomarin/NICP/tree/train/training/data_preprocessing I will try to add more details about...

Hi @adityamurugan , From your pip freeze, I can see that not all the library versions match the expectations. For example, your `open3d` has version `0.18.0`, while it should be`...

Hi @boqian-li , Thanks for taking the initiative. I was going to suggest a similar approach: evaluating CAPE requires similar pipeline described here for FAUST: running the "evaluation_benchmark.py" on the...

By "preparing the data" I meant (in the case your goal is to fully replicate the quantitative evaluation) taking the same frames and test subjects provided in PTF code. Apart...