-
OS: Windows
Python: 3.8
CUDA: 10.2
Torch: 1.7.0
TorchVision: 0.8.1
Reproduce:
python inference/swap.py 1.mp4 -t 2.mp4 --gpus 0 -b 1 -pb 1 -db 1 -lb 1 -sb 1
Error:
0%| | 0/300 [0…
-
Could you please explain (or ideally upload the scripts) how to apply FlashAvatar for face reenactment task?
Should I train FlashAvatar on source head, run tracking on the target head, and then in …
-
1. As for 3D face tracking, whether it is obtained from the original image(original video is 1920*1080)and whether it is affected by camera calibration, If I using deca to calculate the 3D key points …
-
Hi, dear author. I have some questions about self-reenactment and cross-identity reenactment.
1. For multi-view, the FLAME meshes have vertices at varied positions but share the same topology. Does…
-
How to test the Audio Driven?
-
### Describe the problem
Dramatic reenactment of a potential future Discord thread:
> Woot plugins! Woot packages! Oh wait, more packages. Errrrmmm what version goes with what? Do I have the righ…
-
# TODO
- [x] Track head pose.
- [x] Improve virtual scene.
# References
- [MediaPipe-Face-Mesh metric-3d](https://github.com/google-ai-edge/mediapipe/wiki/MediaPipe-Face-Mesh#metric-3d-space).…
-
I used anaconda virtual environment and also .venv python 3.10 - tested on both. I also tested it in python 3.11.
When installed dependencies i tried to make inference for Face Reenactment and got th…
A-2-H updated
1 month ago
-
请问论文接受后,会放出训练代码吗?
-
So I feel like Winnie made a great description of your mini_ex, so I will refrain from trying to elaborate further on a reenactment of the very excellent program that you have written. The effort that…