Pixel2Mesh

Pixel2Mesh copied to clipboard

ValueError: You need to provide a valid (batch_size)x3x4 projection matrix

3

After the solved installation problem with the use of a docker image based on CUDA 10 and some changes on the cpp code of the neural renderer the code start,...

marcomameli1992

CUDA 11 could be supported by replacing .gitmodules url by different n3mr repository

1

I got this repo working in following library versions: * Ubuntu 20 * RTX 3090 * Python 3.7 * PyTorch 1.7.0 * CUDA 11.0 * OpenCV 4.1 * Scipy 1.3...

kimjson

Question on F.grid_sample for mesh to pixel

1

Hi there, Thanks for providing the code for pixel2mesh, nice work! I have one question for computing the 2D image features by computing the correspondence between pixel and vertex in...

YiruS

Confused on training data

1

Thanks a lot for releasing this code. May I know if you could provide a simple dataset for running the code? Just for making reading the code more easily. My...

YanhaoZhang

How to decompress the data after downloading.

1

How to decompress the data after downloading. thank you.

yeyuxmf

Error in DataParallel when trying to train with multiple gpus with torch 1.11

1

I am getting this error when trying to train with multiple gpus: File ./lib/python3.7/site-packages/torch/nn/parallel/replicate.py", line 71, in _broadcast_coalesced_reshape NotImplementedError: Could not run 'aten::view' with arguments from the 'SparseCUDA' backend. This...

amirbarda

Configuration for best performance

1

Hi, Could you please share that which configuration would lead to the best performance as mentioned in the results table? Thanks.

ChengChen2020

Questiion about dataset and loss ?

1

Hi, thanks for sharing the repo. I'm doing the research on 3D reconstruction from single image now, and I found your repo. and I have seen that in `Shapenet.py`, you...

brian3021

The checkpoint file that you provide in google drive seems to be damaged and cannot be decompressed.

1

I have downloaded the checkpoint tar file `resnet.pth.tar` and `tensorflow.pth.tar ` from google driver, but they can't be de decompressed:  I have also tried to use some decompression software...

XJTUmuxin

Is the adj_mat constant in all shapes(airplane, car, ellipsoid)?

1

I noticed that the adj_mat is the edges weight between nodes, and they are float numbers. Is the adj_mat constant in every gconv and is suitable for every shape？

L-Aidan