FaceFusion

https://docs.facefusion.io
Admin
Industry leading face manipulation platform for advanced face swapping and editing.
Tokens:29,089
Snippets:561
Trust Score:6.9
Update:4 months ago
Show doc for...
Tokens:
Raw
### Query Documentation Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Demonstrates how to query the documentation dynamically by asking a question via an HTTP GET request.

```http
GET https://docs.facefusion.io/usage/cli-arguments/face-selector.md?ask=<question>
```

--------------------------------

### Query Documentation API Example

Source: https://docs.facefusion.io/usage/cli-arguments/uis.md

Perform an HTTP GET request to query the documentation dynamically. Append the `ask` query parameter with your question to retrieve specific information.

```http
GET https://docs.facefusion.io/usage/cli-arguments/uis.md?ask=<question>
```

--------------------------------

### Install Git on Windows

Source: https://docs.facefusion.io/installation/platform/windows.md

Use this command to install Git via winget.

```bash
winget install -e --id Git.Git
```

--------------------------------

### Install FFmpeg on Windows

Source: https://docs.facefusion.io/installation/platform/windows.md

Installs FFmpeg version 7.0.2 using winget.

```bash
winget install -e --id Gyan.FFmpeg --version 7.0.2
```

--------------------------------

### Specify Start Frame

Source: https://docs.facefusion.io/usage/cli-arguments/frame-extraction.md

Use `--trim-frame-start` to set the initial frame for video processing. Example: `--trim-frame-start 100`.

```bash
--trim-frame-start 100
```

--------------------------------

### Install FaceFusion (OpenVINO)

Source: https://docs.facefusion.io/installation.md

Installs FaceFusion with OpenVINO support. This is optimized for Intel hardware.

```bash
python install.py --onnxruntime openvino
```

--------------------------------

### Install Git on macOS

Source: https://docs.facefusion.io/installation/platform/macos.md

Use Homebrew to install the Git version control system.

```bash
brew install git
```

--------------------------------

### Face Selector Age Start Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Filters detected faces to include only those starting from age 20. Note: The example in the source uses `--trim-frame-start` which appears to be a typo and should likely be `--face-selector-age-start`.

```bash
--trim-frame-start 20
```

--------------------------------

### Install Miniconda on macOS

Source: https://docs.facefusion.io/installation/platform/macos.md

Use Homebrew to install Miniconda, a minimal installer for Conda.

```bash
brew install miniconda
```

--------------------------------

### Query Documentation with GET Request

Source: https://docs.facefusion.io/usage/cli-commands/job-manager.md

Perform an HTTP GET request to query the documentation dynamically. Include your question as the 'ask' query parameter.

```bash
GET https://docs.facefusion.io/usage/cli-commands/job-manager.md?ask=<question>
```

--------------------------------

### Install FaceFusion (CPU)

Source: https://docs.facefusion.io/installation.md

Installs FaceFusion using the CPU backend with onnxruntime. Use this for systems without dedicated GPUs.

```bash
python install.py --onnxruntime default
```

--------------------------------

### Install Miniconda3 on Windows

Source: https://docs.facefusion.io/installation/platform/windows.md

Installs Miniconda3 with Python 3.12 and adds it to the PATH.

```bash
winget install -e --id Anaconda.Miniconda3 --version py312_25.1.1-2 --override "/AddToPath=1"
```

--------------------------------

### Query Documentation via HTTP GET

Source: https://docs.facefusion.io/installation/platform/macos.md

Perform an HTTP GET request to query the documentation dynamically. Append your question as the 'ask' query parameter.

```http
GET https://docs.facefusion.io/installation/platform/macos.md?ask=<question>
```

--------------------------------

### Example: Specify Source Paths

Source: https://docs.facefusion.io/usage/cli-arguments/paths.md

Use the `-s` or `--source-paths` argument to specify one or more image or audio source files.

```bash
--source-paths source.jpg source.mp3
```

--------------------------------

### Example: Specify Target Path

Source: https://docs.facefusion.io/usage/cli-arguments/paths.md

Use the `-t` or `--target-path` argument to specify the output image or video file.

```bash
--target-path target.mp4
```

--------------------------------

### Query Documentation Dynamically

Source: https://docs.facefusion.io/troubleshooting/windows.md

To get more information not explicitly on the page, perform an HTTP GET request with the 'ask' query parameter. The question should be specific and in natural language.

```bash
GET https://docs.facefusion.io/troubleshooting/windows.md?ask=<question>
```

--------------------------------

### Querying Documentation via HTTP GET

Source: https://docs.facefusion.io/usage/cli-arguments/face-detector.md

To get information not explicitly present on a page, perform an HTTP GET request to the page URL with the `ask` query parameter. The question should be specific and self-contained.

```http
GET https://docs.facefusion.io/usage/cli-arguments/face-detector.md?ask=<question>
```

--------------------------------

### Query Documentation via HTTP GET

Source: https://docs.facefusion.io/usage/cli-arguments/processors/frame-enhancer.md

To get answers to specific questions not directly on the page, perform an HTTP GET request with the 'ask' query parameter to the page URL.

```http
GET https://docs.facefusion.io/usage/cli-arguments/processors/frame-enhancer.md?ask=<question>
```

--------------------------------

### Install Miniconda on Linux

Source: https://docs.facefusion.io/installation/platform/linux.md

Executes the downloaded Miniconda installer script to set up Conda environment management on Linux.

```bash
bash Miniconda3-latest-Linux-x86_64.sh
```

--------------------------------

### Example: Specify Output Path

Source: https://docs.facefusion.io/usage/cli-arguments/paths.md

Use the `-o` or `--output-path` argument to specify the final output image or video file within a directory.

```bash
--output-path output.mp4
```

--------------------------------

### Query Documentation Dynamically

Source: https://docs.facefusion.io/introduction/changelog.md

Use this GET request to ask questions about the documentation. The response includes direct answers and relevant excerpts.

```http
GET https://docs.facefusion.io/introduction/changelog.md?ask=<question>
```

--------------------------------

### Install Git on Linux

Source: https://docs.facefusion.io/installation/platform/linux.md

Installs the git-all package, which includes Git and related utilities, on Debian-based Linux distributions.

```bash
apt install git-all
```

--------------------------------

### Query Documentation Dynamically

Source: https://docs.facefusion.io/usage/cli-arguments/benchmark.md

To get information not explicitly on the page, perform an HTTP GET request with the `ask` query parameter. The response includes answers and relevant excerpts.

```http
GET https://docs.facefusion.io/usage/cli-arguments/benchmark.md?ask=<question>
```

--------------------------------

### Query Documentation with GET Request

Source: https://docs.facefusion.io/llms.txt

Use this GET request to query the documentation index with a specific question. The response will include a direct answer and relevant excerpts.

```http
GET https://docs.facefusion.io/troubleshooting.md?ask=<question>
```

--------------------------------

### Query Documentation via HTTP GET

Source: https://docs.facefusion.io/usage/cli-commands/general.md

Use this method to ask questions about the documentation dynamically. The response includes direct answers, excerpts, and sources.

```bash
GET https://docs.facefusion.io/usage/cli-commands/general.md?ask=<question>
```

--------------------------------

### Query Documentation via HTTP GET Request

Source: https://docs.facefusion.io/usage/cli-arguments/download.md

To ask questions about the documentation dynamically, perform an HTTP GET request to the page URL with an 'ask' query parameter. The question should be specific and in natural language.

```http
GET https://docs.facefusion.io/usage/cli-arguments/download.md?ask=<question>
```

--------------------------------

### Install OpenVINO

Source: https://docs.facefusion.io/installation/accelerator/windows.md

Installs OpenVINO for acceleration on Intel Arc graphics cards. This package is provided by conda-forge.

```bash
conda install conda-forge::openvino=2025.3.0
```

--------------------------------

### Install ROCm Package

Source: https://docs.facefusion.io/installation/accelerator/linux.md

Installs the downloaded AMD ROCm package using dpkg.

```bash
dpkg --install amdgpu-install-7.2.deb
```

--------------------------------

### Install FFmpeg on macOS

Source: https://docs.facefusion.io/installation/platform/macos.md

Use Homebrew to install FFmpeg version 7.0.2.

```bash
brew install ffmpeg@7.0.2
```

--------------------------------

### Install FaceFusion (CUDA)

Source: https://docs.facefusion.io/installation.md

Installs FaceFusion with CUDA support for NVIDIA GPUs. Ensure you have a compatible NVIDIA driver and CUDA toolkit installed.

```bash
python install.py --onnxruntime cuda
```

--------------------------------

### Query Documentation via HTTP GET

Source: https://docs.facefusion.io/usage/cli-arguments.md

To get specific information not found directly on a page, make an HTTP GET request to the page URL with an 'ask' query parameter. This allows for dynamic querying of the documentation.

```http
GET https://docs.facefusion.io/usage/cli-arguments.md?ask=<question>
```

--------------------------------

### Install FFmpeg on Linux

Source: https://docs.facefusion.io/installation/platform/linux.md

Installs FFmpeg, a crucial tool for video and audio processing, on Debian-based Linux distributions.

```bash
apt install ffmpeg
```

--------------------------------

### Download Miniconda Installer

Source: https://docs.facefusion.io/installation/platform/linux.md

Downloads the latest Miniconda installer script for 64-bit Linux systems using cURL.

```bash
curl -LO https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
```

--------------------------------

### Download ROCm Installer

Source: https://docs.facefusion.io/installation/accelerator/linux.md

Downloads the AMD ROCm installation package for Ubuntu Jammy using curl.

```bash
curl --output amdgpu-install-7.2.deb https://repo.radeon.com/amdgpu-install/7.2/ubuntu/jammy/amdgpu-install_7.2.70200-1_all.deb
```

--------------------------------

### Example: Specify Jobs Path

Source: https://docs.facefusion.io/usage/cli-arguments/paths.md

Use the `--jobs-path` argument to specify a directory for storing jobs, overriding the default `.jobs`.

```bash
--jobs-path .jobs
```

--------------------------------

### Install FaceFusion (MIGraphX)

Source: https://docs.facefusion.io/installation.md

Installs FaceFusion with MIGraphX support. This is typically used for AMD GPUs on Linux.

```bash
python install.py --onnxruntime migraphx
```

--------------------------------

### Query Documentation via HTTP GET

Source: https://docs.facefusion.io/donation.md

Perform an HTTP GET request to query the documentation dynamically. Use this when the answer is not explicitly present, for clarification, or to retrieve related sections.

```http
GET https://docs.facefusion.io/donation.md?ask=<question>
```

--------------------------------

### Install cURL on Linux

Source: https://docs.facefusion.io/installation/platform/linux.md

Installs the cURL utility, used for transferring data with URLs, on Debian-based Linux distributions.

```bash
apt install curl
```

--------------------------------

### Query Documentation via HTTP GET Request

Source: https://docs.facefusion.io/usage/cli-arguments/face-landmarker.md

To ask questions about the documentation or retrieve additional information, perform an HTTP GET request to the page URL with the 'ask' query parameter.

```http
GET https://docs.facefusion.io/usage/cli-arguments/face-landmarker.md?ask=<question>
```

--------------------------------

### Query Documentation via HTTP GET

Source: https://docs.facefusion.io/usage/cli-commands/job-runner.md

You can dynamically query the documentation by sending an HTTP GET request to the page URL with an 'ask' query parameter. This is useful for retrieving specific information not explicitly present on the page.

```http
GET https://docs.facefusion.io/usage/cli-commands/job-runner.md?ask=<question>
```

--------------------------------

### Query Documentation via HTTP GET

Source: https://docs.facefusion.io/usage/deepfake-webcam.md

Use this method to ask questions about the documentation dynamically. Replace <question> with your natural language query.

```http
GET https://docs.facefusion.io/usage/deepfake-webcam.md?ask=<question>
```

--------------------------------

### Install FaceFusion (DirectML)

Source: https://docs.facefusion.io/installation.md

Installs FaceFusion with DirectML support for Windows. This is an option for DirectX 12 compatible GPUs.

```bash
python install.py --onnxruntime directml
```

--------------------------------

### Example: Specify Config Path

Source: https://docs.facefusion.io/usage/cli-arguments/paths.md

Use the `--config-path` argument to specify a custom configuration file, overriding the default `facefusion.ini`.

```bash
--config-path facefusion.ini
```

--------------------------------

### Face Parser Model Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-masker.md

Select the model for region parsing. Available options include different ResNet backbones.

```bash
--face-parser-model bisenet_resnet_18
```

--------------------------------

### Example: Specify Temp Path

Source: https://docs.facefusion.io/usage/cli-arguments/paths.md

Use the `--temp-path` argument to specify a directory for temporary resources, overriding the default `TemporaryDirectory`.

```bash
--temp-path .temp
```

--------------------------------

### Face Debugger Items CLI Argument Example

Source: https://docs.facefusion.io/usage/cli-arguments/processors/face-debugger.md

Use this argument to specify which face detector and face masker items to visualize. For example, to visualize only the bounding box.

```bash
--face-debugger-items bounding-box
```

--------------------------------

### Face Occluder Model Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-masker.md

Specify the model for occlusion masking. Use 'many' for broader coverage or specific 'xseg' models for targeted occlusion.

```bash
--face-occluder-model xseg_2
```

--------------------------------

### Face Mask Types Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-masker.md

Define the types of masks to be applied. Multiple types can be combined.

```bash
--face-mask-types occlusion
```

--------------------------------

### Install CUDA Runtime and cuDNN

Source: https://docs.facefusion.io/installation/accelerator/windows.md

Installs the CUDA runtime and cuDNN library, compatible with NVIDIA graphics cards. Ensure your NVIDIA driver is up-to-date.

```bash
conda install nvidia/label/cuda-12.9.1::cuda-runtime nvidia/label/cudnn-9.10.0::cudnn
```

--------------------------------

### Navigate to FaceFusion Directory

Source: https://docs.facefusion.io/installation.md

Changes the current directory to the cloned 'facefusion' repository. This is necessary before installation.

```bash
cd facefusion
```

--------------------------------

### Install TensorRT

Source: https://docs.facefusion.io/installation/accelerator/windows.md

Installs TensorRT for high-performance inference on NVIDIA graphics cards. This is suitable for demanding graphical applications.

```bash
pip install tensorrt==10.12.0.36 --extra-index-url https://pypi.nvidia.com
```

--------------------------------

### Query Documentation Dynamically

Source: https://docs.facefusion.io/introduction/disclaimer.md

To get specific information not found on a page, make an HTTP GET request to the page URL with the 'ask' query parameter. The question should be clear and in natural language.

```http
GET https://docs.facefusion.io/introduction/disclaimer.md?ask=<question>
```

--------------------------------

### Specify Download Providers via CLI

Source: https://docs.facefusion.io/usage/cli-arguments/download.md

Use the `--download-providers` argument to select specific providers for downloading. For example, to only use Hugging Face, set it to `huggingface`.

```bash
--download-providers huggingface
```

--------------------------------

### Face Selector Order Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Specifies the order in which detected faces should be processed, from left to right.

```bash
--face-selector-order left-right
```

--------------------------------

### Face Selector Mode Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Sets the face selector mode to 'one', which means only one face will be selected.

```bash
--face-selector-mode one
```

--------------------------------

### Reference Face Position Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Sets the position used to create the reference face to the first detected face (index 1).

```bash
--reference-face-position 1
```

--------------------------------

### Run CPU Docker Container

Source: https://docs.facefusion.io/usage/run-with-docker.md

Start the FaceFusion Docker container optimized for CPU processing. Access the application via the specified local URL.

```bash
docker compose -f docker-compose.cpu.yml up
```

--------------------------------

### Clone FaceFusion Docker Repository

Source: https://docs.facefusion.io/usage/run-with-docker.md

Clone the official FaceFusion Docker repository to your local machine to begin the setup process.

```bash
git clone https://github.com/facefusion/facefusion-docker.git
```

--------------------------------

### Run TensorRT Docker Container

Source: https://docs.facefusion.io/usage/run-with-docker.md

Start the FaceFusion Docker container utilizing TensorRT for optimized NVIDIA GPU inference. Connect to the application using the designated local URL.

```bash
docker compose -f docker-compose.tensorrt.yml up
```

--------------------------------

### Create UI Component File

Source: https://docs.facefusion.io/workshop/ui-component.md

Create a new Python file for your UI component in the specified directory.

```bash
facefusion/uis/components/example.py
```

--------------------------------

### Face Mask Regions Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-masker.md

Choose specific facial regions for the region mask. A wide range of regions are available, from skin to specific facial features.

```bash
--face-mask-regions nose
```

--------------------------------

### Face Mask Areas Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-masker.md

Specify the areas of the face to be used for the area mask. Options include upper-face, lower-face, and mouth.

```bash
--face-mask-areas mouth
```

--------------------------------

### Face Swapper Pixel Boost CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/processors/face-swapper.md

The `--face-swapper-pixel-boost` argument determines the resolution for the face swapper's output. This example sets the boost to '512x512'.

```bash
--face-swapper-pixel-boost 512x512
```

--------------------------------

### Face Mask Padding Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-masker.md

Apply padding to the box mask. Values represent top, right, bottom, and left padding respectively. Two values specify top/bottom and left/right.

```bash
--face-mask-padding 10 20
```

--------------------------------

### Face Swapper Weight CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/processors/face-swapper.md

Specify the `--face-swapper-weight` to control the intensity of the face swap. This example sets the weight to the maximum value of 1.0.

```bash
--face-swapper-weight 1.0
```

--------------------------------

### Install GCC for ROCm

Source: https://docs.facefusion.io/installation/accelerator/linux.md

Installs GCC version 15.2.0 from conda-forge, which may be required for ROCm.

```bash
conda install conda-forge::gcc=15.2.0
```

--------------------------------

### Query Documentation Dynamically

Source: https://docs.facefusion.io/installation/accelerator.md

Use this method to ask questions about the documentation. The response includes direct answers and relevant excerpts.

```http
GET https://docs.facefusion.io/installation/accelerator.md?ask=<question>
```

--------------------------------

### Query Documentation Dynamically

Source: https://docs.facefusion.io/installation/platform.md

Use this method to ask questions about the documentation. The response includes direct answers and relevant excerpts.

```http
GET https://docs.facefusion.io/installation/platform.md?ask=<question>
```

--------------------------------

### Select Lip Syncer Model via CLI

Source: https://docs.facefusion.io/usage/cli-arguments/processors/lip-syncer.md

Use the `--lip-syncer-model` argument to specify which lip-syncing model to use. Available choices include 'edtalk_256', 'wav2lip_96', and 'wav2lip_gan_96'.

```bash
--lip-syncer-model edtalk_256
```

--------------------------------

### Run Platform Benchmark

Source: https://docs.facefusion.io/usage/platform-benchmark.md

Execute the platform benchmark using the facefusion.py script with the 'run' command and 'benchmark' UI layout.

```bash
python facefusion.py run --ui-layouts benchmark
```

--------------------------------

### UI Layouts CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/uis.md

Specify one or more UI layouts to launch using the `--ui-layouts` argument. Available layouts include `benchmark`, `default`, `jobs`, and `webcam`.

```bash
--ui-layouts default jobs
```

--------------------------------

### Face Selector Race Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Filters detected faces to include only individuals of white race. Note: The example in the source uses `--face-selector-gender` which appears to be a typo and should likely be `--face-selector-race`.

```bash
--face-selector-gender white
```

--------------------------------

### Initialize Conda

Source: https://docs.facefusion.io/installation.md

Initializes conda for your terminal. Run this command before creating environments.

```bash
conda init --all
```

--------------------------------

### Face Selector Age End Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Filters detected faces to include only those up to age 40. Note: The example in the source uses `--trim-frame-end` which appears to be a typo and should likely be `--face-selector-age-end`.

```bash
--trim-frame-end 40
```

--------------------------------

### Face Selector Gender Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Filters detected faces to include only males.

```bash
--face-selector-gender male
```

--------------------------------

### Activate Conda Environment

Source: https://docs.facefusion.io/installation.md

Activates the 'facefusion' conda environment. This should be done before installing or running FaceFusion.

```bash
conda activate facefusion
```

--------------------------------

### Reference Frame Number Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Sets the frame number from which to extract the reference face to frame 100.

```bash
--reference-frame-number 100
```

--------------------------------

### Reference Face Distance Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-selector.md

Specifies the similarity threshold between the reference face and target faces to 0.6.

```bash
--reference-face-distance 0.6
```

--------------------------------

### Set Benchmark Mode

Source: https://docs.facefusion.io/usage/cli-arguments/benchmark.md

Use the `--benchmark-mode` argument to select the benchmark execution mode. Available choices are `warm` and `cold`.

```bash
--benchmark-mode cold
```

--------------------------------

### Run Deepfake Webcam UI

Source: https://docs.facefusion.io/usage/deepfake-webcam.md

Execute the FaceFusion script to run the webcam interface. Ensure you are in the correct directory.

```bash
python facefusion.py run --ui-layouts webcam
```

--------------------------------

### Specify Download Scope via CLI

Source: https://docs.facefusion.io/usage/cli-arguments/download.md

Use the `--download-scope` argument to define the extent of the download. Setting it to `full` will download all available components.

```bash
--download-scope full
```

--------------------------------

### Open Browser CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/uis.md

Use the `--open-browser` argument to automatically open the browser once the program is ready. This is useful for immediate access to the UI.

```bash
--open-browser
```

--------------------------------

### UI Workflow CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/uis.md

Choose the desired UI workflow with the `--ui-workflow` argument. Options are `instant_runner`, `job_runner`, and `job_manager`.

```bash
--ui-workflow job_manager
```

--------------------------------

### Run Program with Custom UI Layout

Source: https://docs.facefusion.io/workshop/ui-layout.md

Execute the FaceFusion program with your newly created UI layout by specifying its name using the `--ui-layouts` argument.

```bash
python run.py --ui-layouts example
```

--------------------------------

### Set Output Video Preset

Source: https://docs.facefusion.io/usage/cli-arguments/output-creation.md

Balance video processing speed and file size by selecting a video preset. Options range from 'ultrafast' for speed to 'veryslow' for better compression.

```bash
--output-video-preset ultrafast
```

--------------------------------

### Specify End Frame

Source: https://docs.facefusion.io/usage/cli-arguments/frame-extraction.md

Use `--trim-frame-end` to define the final frame for video processing. Example: `--trim-frame-end 200`.

```bash
--trim-frame-end 200
```

--------------------------------

### Keep Temporary Files

Source: https://docs.facefusion.io/usage/cli-arguments/frame-extraction.md

Use the `--keep-temp` argument to retain temporary frame resources after processing is complete. Example: `--keep-temp`.

```bash
--keep-temp
```

--------------------------------

### Set Face Landmarker Model via CLI

Source: https://docs.facefusion.io/usage/cli-arguments/face-landmarker.md

Use the `--face-landmarker-model` argument to specify the desired face landmark detection model. Available choices include 'many', '2dfan4', and 'peppa_wutz'.

```bash
--face-landmarker-model peppa_wutz
```

--------------------------------

### Face Mask Blur Example

Source: https://docs.facefusion.io/usage/cli-arguments/face-masker.md

Control the blur intensity applied to the box mask. The value ranges from 0 (no blur) to 1 (maximum blur).

```bash
--face-mask-blur 0.5
```

--------------------------------

### Face Swapper Model CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/processors/face-swapper.md

Use the `--face-swapper-model` argument to select the specific model for face swapping. This example selects the 'hyperswap_1a_256' model.

```bash
--face-swapper-model hyperswap_1a_256
```

--------------------------------

### Create New UI Layout File

Source: https://docs.facefusion.io/workshop/ui-layout.md

To create a new UI layout, generate a Python file within the `facefusion/uis/layouts/` directory. This serves as the entry point for your custom UI.

```bash
facefusion/uis/layouts/example.py
```

--------------------------------

### Select Background Remover Model

Source: https://docs.facefusion.io/usage/cli-arguments/processors/background-remover.md

Use the `--background-remover-model` argument to specify which model to use for background removal. Available choices include various `ben`, `birefnet`, `corridor_key`, `isnet`, `modnet`, `rmbg`, `silueta`, and `u2net` models.

```bash
--background-remover-model rmbg_2.0
```

--------------------------------

### Specify Execution Providers via CLI

Source: https://docs.facefusion.io/usage/cli-arguments/execution.md

Use the `--execution-providers` argument to select the inference provider. Common choices include `cuda`, `cpu`, and `directml`.

```bash
--execution-providers cuda
```

--------------------------------

### Default Overrides for Face Swapper

Source: https://docs.facefusion.io/usage/cli-arguments/processors/face-swapper.md

Modify the `facefusion.ini` file under the `[processors]` section to set default values for face swapper arguments. This example overrides the model, pixel boost, and weight.

```ini
[processors]
face_swapper_model = hyperswap_1a_256
face_swapper_pixel_boost = 512x512
face_swapper_weight = 1.0
```

--------------------------------

### Override Download Defaults in Configuration File

Source: https://docs.facefusion.io/usage/cli-arguments/download.md

Modify the `facefusion.ini` file under the `[download]` section to set default values for download providers and scope. This example sets both to `huggingface` and `full` respectively.

```ini
[download]
download_providers = huggingface
download_scope = full
```

--------------------------------

### Run CUDA Docker Container

Source: https://docs.facefusion.io/usage/run-with-docker.md

Launch the FaceFusion Docker container with CUDA support for GPU acceleration. The application will be accessible at the provided local URL.

```bash
docker compose -f docker-compose.cuda.yml up
```

--------------------------------

### Implement UI Component Basics

Source: https://docs.facefusion.io/workshop/ui-component.md

Implement the fundamental structure of a UI component using Gradio. This includes defining global variables, rendering the component, and setting up event listeners.

```python
from typing import Optional
import gradio

from facefusion.uis.typing import Update

EXAMPLE_IMAGE : Optional[gradio.Image] = None


def render() -> None:
	global EXAMPLE_IMAGE

	EXAMPLE_IMAGE = gradio.Image()


def listen() -> None:
	EXAMPLE_IMAGE.change(update, inputs = EXAMPLE_IMAGE, outputs = EXAMPLE_IMAGE)

def update() -> Update:
	return gradio.update()
```

--------------------------------

### Specify Temporary Frame Format

Source: https://docs.facefusion.io/usage/cli-arguments/frame-extraction.md

Use `--temp-frame-format` to select the file format for temporary frames. Supported formats include bmp, jpeg, png, and tiff. Example: `--temp-frame-format bmp`.

```bash
--temp-frame-format bmp
```

--------------------------------

### Create Processor File

Source: https://docs.facefusion.io/workshop/processor.md

Create a new Python file for your processor module within the facefusion/processors/modules/ directory. This is the first step in developing a custom processor.

```bash
facefusion/processors/modules/example.py
```

--------------------------------

### Source Pattern CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/patterns.md

Use the `--source-pattern` argument to specify the pattern for input images or audio files. The default value is None.

```bash
--source-pattern sources\*.jpg
```

--------------------------------

### Specify Processors via CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/processors.md

Use the `--processors` argument to load one or more processors. The default processor is `face_swapper`.

```bash
--processors face_swapper face_enhancer
```

--------------------------------

### Run Processor

Source: https://docs.facefusion.io/workshop/processor.md

Execute the FaceFusion program with your custom processor enabled. Use the --processors argument to specify the name of your processor module.

```bash
python run.py --processors example
```

--------------------------------

### Run FaceFusion Application

Source: https://docs.facefusion.io/installation.md

Executes the FaceFusion program with the 'run' command and automatically opens the browser. Ensure your environment is activated.

```bash
python facefusion.py run --open-browser
```

--------------------------------

### Clone FaceFusion Repository (Gitee)

Source: https://docs.facefusion.io/installation.md

Clones the FaceFusion repository from Gitee. Use this command to download the source code.

```bash
git clone https://gitee.com/facefusion/facefusion
```

--------------------------------

### Add UI Component to Application

Source: https://docs.facefusion.io/workshop/ui-component.md

Import and add the newly created UI component to the application's main structure.

```python
from facefusion.uis.components import example
```

--------------------------------

### Implement UI Layout Hooks

Source: https://docs.facefusion.io/workshop/ui-layout.md

Implement the required hooks (`pre_check`, `render`, `listen`, `run`) for your UI layout. The `render` hook is responsible for defining the UI structure using Gradio components.

```python
import gradio

from facefusion.uis.components import example

def pre_check() -> bool:
	return True

def render() -> gradio.Blocks:
	with gradio.Blocks() as layout:
		example.render()
	return layout

def listen() -> None:
	example.listen()

def run(ui : gradio.Blocks) -> None:
	ui.launch()
```

--------------------------------

### Set Video Memory Strategy CLI Argument

Source: https://docs.facefusion.io/usage/cli-arguments/memory.md

Use the --video-memory-strategy argument to control how FaceFusion balances fast frame processing and low VRAM usage. Options include 'strict', 'moderate', and 'tolerant'.

```bash
--video-memory-strategy tolerant
```

--------------------------------

### Set Output Video Scale

Source: https://docs.facefusion.io/usage/cli-arguments/output-creation.md

Specify the video scale relative to the target video. Use this argument to resize the output video.

```bash
--output-video-scale 0.5
```

--------------------------------

### Set Benchmark Resolutions

Source: https://docs.facefusion.io/usage/cli-arguments/benchmark.md

Specify desired resolutions for benchmark tests using the `--benchmark-resolutions` argument. Multiple resolutions can be provided.

```bash
--benchmark-resolutions 240p 360p
```

--------------------------------

### Set Lip Syncer Weight via CLI

Source: https://docs.facefusion.io/usage/cli-arguments/processors/lip-syncer.md

Adjust the `--lip-syncer-weight` argument to control the influence of the lip syncing. The weight can range from 0.0 to 1.0, with a default of 0.5.

```bash
--lip-syncer-weight 1.0
```

--------------------------------

### Set Frame Colorizer Model via CLI

Source: https://docs.facefusion.io/usage/cli-arguments/processors/frame-colorizer.md

Use the --frame-colorizer-model argument to specify the desired colorization model. Available models include ddcolor, ddcolor_artistic, deoldify, deoldify_artistic, and deoldify_stable.

```bash
--frame-colorizer-model deoldify
```
Tokens:
Raw