Zheng Ding

Could you provide more details on how you get the lambertian rendering and what is the exact image you're encountering the problem? It seems fine on my end.

Since we utilize DECA to obtain physical buffers but DECA does not output eye_pose parameters, the model could not learn to control the eye pose. FLAME would use a default...

We can provide the images for Obama we used. Please download it here: https://drive.google.com/file/d/1LJxzacKy_pPurHpfpER1zz7lfnOICj-e/view?usp=sharing

We've released the stage one model in the following link: https://drive.google.com/file/d/1lnFLNGguvQ150unuOXJqbXiVQtJz0jli/view?usp=sharing

The meanshape is used for some cases that the overall shape is hard to estimate such as side faces. You'd better use some personal photos to finetune your model to...

The pose parameters represent both neck and jaw poses. The last three are for jaw rotations, therefore we include them in the expression. You can see the figure 2 in...

Hi, in order to edit the expression directly without the condition image, you can try to modify the parameters here https://github.com/adobe-research/diffusion-rig/blob/d9b266102e1900032efc62e66a7f6c66dde5ff79/scripts/inference.py#L80C1-L81C1 where you directly change original expression parameters instead of...

Hi, thanks for your interest in our work! Yes, I think that should be feasible. Ideally any model like DECA might work. We choose DECA since it's easy to use...

I didn't encounter this issue before. You mentioned that the problem happened in the sync_params(), can you try to remove this for a quick workaround? Since the stage 2 only...