Linux下GeneFace++安装指南

Note: The following guide is compiled from community-contributed articles for GeneFace++ installation on CentOS (a common Linux distribution). Since GeneFace++ is not a widely documented open-source project, always refer to the official GitHub repository or project documentation for the most accurate and up-to-date instructions.

1. Preparation

Before starting, ensure your Linux system (e.g., CentOS 7/8) is updated and has basic development tools installed. A stable internet connection is required to download dependencies and source code.
Key Prerequisites:

• Git (for cloning repositories)
• CMake (for building from source)
• Python 3.9+ (if the project includes Python components)
• GPU with CUDA support (if using hardware acceleration, ensure CUDA/cuDNN versions match project requirements).

2. Update System and Install Core Dependencies

Update your system to the latest package versions and install essential development tools:

sudo yum update -y                  # Update all system packages
sudo yum groupinstall -y "Development Tools"  # Install GCC, Make, etc.
sudo yum install -y cmake git wget  # Install CMake (build tool), Git (version control)

For projects requiring Python or image processing libraries, add these dependencies:

sudo yum install -y python3 python3-devel libpng-devel libjpeg-devel libtiff-devel  # Python and image libraries

3. Clone GeneFace++ Source Code

Replace the placeholder repository URL with the actual GeneFace++ GitHub address (verify this from the official project page):

git clone https://github.com/your-repo/GeneFacePlusPlus.git  # Replace with real repo
cd GeneFacePlusPlus  # Enter the project directory

4. Build and Install from Source

Most GeneFace++ projects use CMake for compilation. Follow these steps:

mkdir build && cd build  # Create a build directory (recommended for out-of-source builds)
cmake ..                 # Generate Makefiles (checks dependencies and configures build)
make -j$(nproc)          # Compile the project (uses all available CPU cores for faster builds)
sudo make install        # Install binaries/libraries to system directories (e.g., /usr/local/bin)

5. Configure Environment Variables (Optional but Recommended)

If you installed GeneFace++ to a non-standard directory (not /usr/local), add its bin (for executables) and lib (for libraries) paths to your shell environment:

echo 'export PATH=$PATH:/path/to/GeneFacePlusPlus/build/bin' >> ~/.bashrc  # Add binaries to PATH
echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/path/to/GeneFacePlusPlus/build/lib' >> ~/.bashrc  # Add libraries to LD_LIBRARY_PATH
source ~/.bashrc  # Apply changes to the current terminal session

6. Verify Installation

Run the project’s help command or test script to confirm successful installation. For example:

geneface++ --help  # Display command-line options (replace with actual executable name)

If the project includes Python scripts (e.g., for inference), activate any required Python environments and run a demo:

# Example for Python-based inference (adjust paths as needed)
conda activate geneface  # Activate conda environment (if used)
python inference/genefacepp_infer.py --a2m_ckpt=checkpoints/audio2motion_vae --drv_aud=data/raw/val_wavs/example.wav --out_name=demo_output.mp4

7. Optional: Run as a System Service (for Production)

To run GeneFace++ as a background service (e.g., for continuous operation), create a systemd service file:

sudo nano /etc/systemd/system/geneface++.service  # Open the file in a text editor

Add the following content (modify paths/user/group as needed):

[Unit]
Description=GeneFace++ Service
After=network.target
[Service]
ExecStart=/path/to/GeneFacePlusPlus/build/bin/geneface++
Restart=always
User=your_username  # Replace with your Linux username
Group=your_groupname  # Replace with your Linux groupname
[Install]
WantedBy=multi-user.target

Save the file, then start and enable the service:

sudo systemctl daemon-reload  # Reload systemd to recognize the new service
sudo systemctl start geneface++  # Start the service
sudo systemctl enable geneface++  # Enable auto-start on boot

Check logs to monitor runtime status:

sudo journalctl -u geneface++ -f  # Follow real-time logs

Troubleshooting Tips

• Dependency Issues: If cmake fails due to missing libraries, install the required packages (refer to error messages) or configure a domestic Yum mirror (e.g., Alibaba Cloud) to speed up downloads.
• Python Environment Conflicts: Use conda or venv to isolate Python dependencies and avoid conflicts with system packages.
• Build Failures: Ensure your GPU drivers/CUDA versions match the project requirements (check the README for details).

For further assistance, consult the official GeneFace++ documentation or community forums (if available).