Hello, first of all, thank you so much for your open source. I am very interested in your work.
I have finished the finetuning exm in my personal album flowing your Redme.md. The result remains good and fantastic! But I have a question: as stated in your paper, the diff-org will be affected by DECA’s limited estimation capability. So can I replace the DECA part with Deep3D model? Do you have any suggestions for me if I do the replacement?
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 and manipulate.
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 and manipulate.
Hi, I encountered this issue while running the project and would like to ask you how to solve it.
creating the FLAME Decoder
trained model found. load /hy-tmp/2023A/DECA/diffusion-rig-main/data/deca_model.tar
/usr/local/miniconda3/envs/diffusionrig1.11/lib/python3.8/site-packages/pytorch3d/io/obj_io.py:546: UserWarning: Mtl file does not exist: /hy-tmp/2023A/DECA/diffusion-rig-main/data/template.mtl
warnings.warn(f"Mtl file does not exist: {f}")
Traceback (most recent call last):
File "scripts/create_data.py", line 133, in
I met the sample problem
@QvQKing you can fix it according to solution in https://stackoverflow.com/questions/76651887/am-facing-attributeerror2d-how-do-i-resolve-it