fanziapril/mvfnet

Pytorch code for paper: MVF-Net: Multi-View 3D Face Morphable Model Regression

mvfnet's Issues

no modelshape mat

I cannot locate the model shape mat file it is not in the 3ddfa zip

Differentiable Rendering

Thank you for your job.
How to do Differentiable Rendering？Use https://github.com/google/tf_mesh_renderer？

how to project the model to front image?

Hi,
Thanks for your excellent work. I want to project the model to the front image of three input images, could you please tell me how to project the model to front image?

Thanks!

how to calculate the distance between reconstrution and groundtruth?

teeth is not covered by the mesh

Model_shape.mat

Hello, I download the code of 3DDFA, but there is only the file Model_Expression.mat. May I ask where the file Model_shape.mat is ?

Thankyou~

No Texture

No texture found after construction

the .obj format output of test_img.py can't show face color information in meshalb

Maybe it's a stupid question...

how to get a frontal rendered image?

@Fanziapril, Hi, I am appreciated with your great job? but I have some question.
How could I get a absolute frontal rendered image given the generated 3d face model and a InputB image which is a relative frontal face?

result of demo is not accordance with that in paper

Can anyone provide a download link for modelshape.mat?

modelshape.mat is hard to get

what should I do if I want to input more images ?

I can't find Model_shape.mat in 3DDFA.

only Model_Expression.mat in 3DDFA.
what should I do?

CNN model

Sorry,For some reason(vpn is error...0_0),I can't download CNN model in dropbox,and could you save it in Baiduyun or other address？

Thanks for your great work, but i can't find Model_shape.mat in 3DDFA and i can't download net.pth by dropbox, can you give me some help?

about error map comparison on MICC

Hello, thanks for your great work, I want to know how to do the experiments like Figure 7 in your paper, can you give me some advice or share your test code?
Thanks a lot!

Hello, which torchvision I should install for MVF NET?

When I tried to install with pip install torchvision, it kept deleting pytorch 0.4.0 and installing recent version.

Performance on florence is inconsistent with the paper

Hi. I've tried your model on florence dataset and do a quantitative experiment. However, in my test your method is worse than Genova et al.. I wonder if there is some limitations i did not notice. For example, do you allow left and right images with large poses (yaw angles larger than 60 degrees) ? By the way, florence dataset has multiple images for a single subject under different scenes, do you use only 3 images of a subject for evaluation in your paper?

textured output

how to get textured output with the ply

can't find 01_MorphableModel.mat

How to install the code:

1. Downloading the Basel Face Model (BFM) on "
   http://faces.cs.unibas.ch/bfm/main.php?nav=1-0&id=basel_face_model"
2. Copy the "01_MorphableModel.mat" file in the BFM to
   Matlab/ModelGeneration/
3. Run the Matlab/ModelGeneration/ModelGenerate.m to generate the shape
   model "Model_Shape.mat" and copy it to the Matlab/

hi,
I can't find 01_MorphableModel.mat in (1),just get BaselFaceModel.

Some thing is wrong with shape.ply

Hi, I run the example but got such result.


May I ask how could I solve this problem ?
Thankyou~

A question about experiment

Hi! Thanks for your great work in 3D face reconstruction!

I have a question about the processing of MICC Florence dataset.

How did you process the scan results, especially the crop step? Because I find it difficult to find the nose point accurately T_T...

pose parameters

Thank you for your job.
I found that there are seven pose parameters. What do they mean respectively?