OpenTalker / SadTalker

[CVPR 2023] SadTalker:Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation
https://sadtalker.github.io/
Other
11.24k stars 2.1k forks source link

3D Face visualization doesnt work #449

Open G-force78 opened 1 year ago

G-force78 commented 1 year ago

Missing many dependencies it seems, I simply copied the code into a collab notebook from here https://github.com/OpenTalker/SadTalker/blob/main/docs/face3d.md

Traceback (most recent call last): File "/content/SadTalker/inference.py", line 8, in from src.utils.preprocess import CropAndExtract File "/content/SadTalker/src/utils/preprocess.py", line 11, in from src.face3d.models import networks File "/content/SadTalker/src/face3d/models/init.py", line 22, in from src.face3d.models.base_model import BaseModel File "/content/SadTalker/src/face3d/models/base_model.py", line 9, in from . import networks File "/content/SadTalker/src/face3d/models/networks.py", line 19, in from kornia.geometry import warp_affine ModuleNotFoundError: No module named 'kornia'

vinthony commented 1 year ago

try pip install korina

G-force78 commented 1 year ago

try pip install korina

Kornia was in the res3d file already, I guess it needs another version? Same for yacs

Ok now it starts after installing these !pip install korina !pip install safetensors !pip install yacs !pip install pydub

but now I get FileNotFoundError: [Errno 2] No such file or directory: './checkpoints/epoch_20.pth'

vinthony commented 1 year ago

have you download all the checkpoints?

G-force78 commented 1 year ago

have you download all the checkpoints?

No I thought it would have already done it, I will try again thanks

G-force78 commented 10 months ago

I now get this error

Traceback (most recent call last): File "/content/SadTalker/inference.py", line 144, in main(args) File "/content/SadTalker/inference.py", line 79, in main from src.face3d.visualize import gen_composed_video File "/content/SadTalker/src/face3d/visualize.py", line 5, in from src.face3d.models.facerecon_model import FaceReconModel File "/content/SadTalker/src/face3d/models/facerecon_model.py", line 11, in from src.face3d.util.nvdiffrast import MeshRenderer File "/content/SadTalker/src/face3d/util/nvdiffrast.py", line 4, in import pytorch3d.ops ModuleNotFoundError: No module named 'pytorch3d'

Have looked back and found this

ERROR: Could not find a version that satisfies the requirement pytorch3d (from versions: none) ERROR: No matching distribution found for pytorch3d