Open aihacker111 opened 1 month ago
Very nice work congrats
Thank you very much for your efforts in making LivePortrait a better project! We will follow up on your thread next week. @aihacker111
rụng trứng mất thôi :v:
@aihacker111 Where can I find the Colab notebook? I want to run this on Google Colab.
@PointerSoftware I'll create notebook for that in this afternoon
@PointerSoftware I'll create notebook for that in this afternoon
Eagerly waiting for that!, Please remember to create a notebook that supports Gradio, so anyone can use the project with a graphical interface.
~期待~
@aihacker111 Thanks for your great work!! I would like to know where can I find the C++ inference code. :))
@ChunChenLin I'm currently still not implement C++ code , maybe in the next week :)
@ChunChenLin I'm currently still not implement C++ code , maybe in the next week :)
@aihacker111 Sure, thanks! Looking forward to it. Thanks again for your great work. :))
@aihacker111 Hey, how's everything :) Just wanna know is the C++ inference code available or coming soon? :))
So during the My Journey to optimize this model run by ONNX GPU and CPU , we got the result like this:
This demo is run on P100 16gb in Kaggle Kernel
This optimiztion will reduce the time speed for run all process of Live-Portrait to 1s/1 frame with 354 frame in total 4 minute 30s
Also support running on Google Colab
Please following my source for run demo: https://github.com/aihacker111/Efficient-Live-Portrait This source already support for run on CPU and Apple silicon , and CUDA
TensorRT models
I'm already converted to TensorRT and in progress building C++ code for inference and demo. Wait the great TensorRT in the next week 🔥🔥🔥🔥🔥 Have fun with result:
https://github.com/user-attachments/assets/0ea322c3-2979-4155-bd58-6c9e3f24e974
https://github.com/user-attachments/assets/010dd312-627a-4488-8dfc-ce7064d92e0a