TimoBolkart
TimoBolkart
Can you provide one Deep3DFace mesh for reference?
No, we did not decay the learning rate and we did not investigate the effect of different learning rates. If you find some better training strategy, I would be happy...
Yes, that is precisely what we did. The BFM template was deformed with some non-rigid ICP to fit the FLAME template.
Hey, given FLAME vertices (output by any method) you can get all FLAME parameters back (i.e., pose, shape, expression) with a simple optimization. You just need to minimize the difference...
All errors reported in the [NoW Challenge list](https://now.is.tue.mpg.de/nonmetricalevaluation.html) are computed on the NoW test set. And yes, DECA reported a larger (mean) error (1.46 mm, see Section 6.3) on the...
There are two common issues that could lead to that, wrongly loading the meshes which can happen if one exports the meshes with vertex colors which the used mesh loader...
Can you please provide some more details about where you replaced the FLAME_texture.npz? We do have some [demo code](https://github.com/TimoBolkart/TF_FLAME/blob/master/sample_texture.py) online to sample textures from FLAME_texture or the albedoMM_texture. Does that...
You can render a depth image from the mesh, I think pytorch3d directly supports this but I have never tried it
Can you provide me mesh that is output by 3DDFA_V2?
Yes, tf_exp are the expression parameters. The Mesh package has a MeshViewer to visualize meshes, otherwise one can output the FLAME mesh and check it in MeshLab