TimoBolkart

Hello, there are different possibilities to get a texture space for FLAME 1. convert the color space of Basel face model (BFM) by running [this code](https://github.com/TimoBolkart/BFM_to_FLAME) 2. download the FLAME...

The provided parameters are specific for FLAME 2019. While FLAME 2020 is able to perform eye blinks, the parameters are not directly compatible.

Hi, the rendering is the same, regardless if you have a texture from the parametric texture models (BFM, AlbedoMM, or the FLAME texture space). The easiest way is to just...

Indeed, changing betas 300-400 can change the expression but it will mess up the speech-related expressions. For expression control, one would need to disentangle the speech- and emotion-related expression to...

You can use FAN, but you need to ignore the first 17 landmarks, which correspond to the face boundary / jaw line of the face. The 51 landmarks represent only...

The FLAME2020 used in EMOCA is compatible with the SMPL-X 2020 used in e.g., [PIXIE](https://pixie.is.tue.mpg.de/). Directly transferring the expression parameters obtained from EMOCA to these SMPL-X parameters should do the...

Hey, for expression, these are the first 50 expression parameters of FLAME, for pose, these are the axis-angle rotations for global rotation (neck) and for jaw. The head, and eyeball...

In this fitting scenario, the model is trying to explain the entire head including hair cap etc. of the target mesh. This can hinder the overall fitting performance, as the...

While this is true, the problem in this case is likely not an optimization problem but a problem of optimizing with too many constraints. In that particular example it seems...

The FLAME shapedirs is a tensor of shape 5023 x 3 x 400, where the first 300 basis vectors are for identity, and the remaining 100 are for expression. In...