stevenlsw/contactgen

ContactGen: Generative Contact Modeling for Grasp Generation (ICCV 2023)

contactgen's Issues

pip install "git+https://github.com/facebookresearch/pytorch3d.git" falied

Dear authors, first of all thank you for your very good work. However, I ran into some problems while trying to reproduce your work.
When I executed "pip install "git+https://github.com/facebookresearch/pytorch3d.git"", I found an error.

note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for pytorch3d
Running setup.py clean for pytorch3d
Failed to build pytorch3d
ERROR: Could not build wheels for pytorch3d, which is required to install pyproject.toml-based projects

May I ask the author, how to solve this problem?

Why can't I reproduce the results from the same seed?

The configurations on the left and right are exactly the same, except that they run for 3 minutes before and after

Dear author,

Hello, I am a first-year graduate student. While attempting to reproduce your work, I encountered an issue where using the same seed does not lead to an exact replication of the previous results. I am wondering why this might be the case. I would greatly appreciate it if you could provide an explanation.

Thank you in advance for your response.

demo.py cannot reproduce the REAME's result

Thank you for your great work. I run the demo.py, following the README, but I got this result. Is there any way to fix this result?

Contact map in inference

Hi,
Can the contact map of a grabbed object be exported in inference?

Thanks

Can the data format use mano parameters?

Dear author,

Greetings. Your work has truly presented an excellent solution. I have noticed that the parameters format used in your team's code differs from that of MANO（61 param）. May I kindly inquire if you would be able to offer insights or guidance on this matter? Your assistance would be immensely appreciated.

Best regards,

The problem about dataset

Dear authors,
First and foremost, I would like to extend my sincere gratitude for sharing your work on GitHub.
I am reaching out to inquire about certain aspects of the dataset used in your published code, specifically regarding the preprocessing of the grabnet dataset. My questions are as follows:
1、In the code that you have made available, I noticed that the preprocessing is done on a subset of the grabnet dataset. Could you please confirm if the comparisons made in your paper also utilized the same subset of the dataset?
2、If the full grabnet dataset was used in your research paper, I would be interested to know the configuration of the training equipment you employed for this purpose.

Your insights on these matters would be incredibly helpful for my understanding and further exploration of your work.
Thank you in advance for your time and response.
Best regards