search

warmshao / FasterLivePortrait

Bring portraits to life in Real Time！onnx/tensorrt support！实时肖像驱动！
521 stars 47 forks source link

easy tutorial on linux #97

Open AbhinavJangra29 opened 2 weeks ago

AbhinavJangra29 commented 2 weeks ago

Installation Guide

For those running into issues with installation, here’s a streamlined guide! This has been tested on the following container:

Steps

  1. Clone the Repository 

    git clone <this-repo-url>
cd <repo-directory>

  2. Install Requirements 

    pip install -r requirements.txt

  3. Install FFmpeg 

    sudo apt-get install ffmpeg

  4. Install ONNX Runtime 

    pip install onnxruntime

Skip the TensorRT Conversion

Note: To avoid building and converting to TensorRT, you can directly download pre-trained weights!

Download the pre-built TensorRT weights using the Hugging Face CLI:

huggingface-cli download mlgawd/trt-weights --local-dir ./checkpoints

Run the Application

Now, you’re ready to run the app in TensorRT mode:

python app.py --mode trt

No need to modify paths or get confused! Just follow these steps and you’re all set. 😎

warmshao commented 2 weeks ago

Thank you, but the TRT weight is associated with the cuda version and NVDIA GPU, so it cannot be used directly like this.

AbhinavJangra29 commented 2 weeks ago

yes for rtx3090 it works

Rizwanali324 commented 2 weeks ago

yes for rtx3090 it works

have you deploy it as inference API at runpod ? what was speed at A10 GPu