Visit ComfyUI Online for ready-to-use ComfyUI environment
Load pre-trained face swap model for AI artists, simplifying face swapping tasks with InsightFace library integration.
The Load Face Swap Model (mtb) node is designed to load a face swap model, enabling you to perform face swapping tasks using deep learning models from the InsightFace library. This node is essential for AI artists who want to create realistic face swaps in their projects. By loading a pre-trained face swap model, you can seamlessly integrate advanced face swapping capabilities into your workflow, allowing for high-quality and accurate face transformations. The node simplifies the process of loading and managing face swap models, ensuring that you can focus on the creative aspects of your work without worrying about the underlying technical complexities.
The faceswap_model
parameter specifies the face swap model to be loaded. This parameter accepts a list of available models, which are dynamically retrieved from the InsightFace model directory. The available options are the names of the models found in the directory with extensions .onnx
or .pth
. The default value is "None". This parameter is crucial as it determines which pre-trained model will be used for the face swapping process. Selecting the appropriate model can significantly impact the quality and accuracy of the face swap results.
The FACESWAP_MODEL
output parameter represents the loaded face swap model. This output is essential as it provides the actual model object that will be used for performing face swaps. The model is an instance of the INSWapper
class, which is initialized with the specified model file and configured to use the available ONNX runtime providers. This output is critical for subsequent nodes that perform the face swapping operation, ensuring they have access to the loaded model.
<model_name>
(<model_path>
)faceswap_model
parameter. Ensure that the file has the correct extension (.onnx or .pth).faceswap_model
parameter. If no models are listed, check that the model files are correctly placed in the InsightFace model directory.© Copyright 2024 RunComfy. All Rights Reserved.