Visit ComfyUI Online for ready-to-use ComfyUI environment
Integrate reference motion data for model output control in AI art and animation tasks.
The ApplyRefMotionNode is designed to integrate reference motion data into a model, enhancing its ability to generate outputs that adhere to specific motion patterns or characteristics. This node is particularly useful for AI artists who want to apply motion references to their models, ensuring that the generated outputs follow desired motion dynamics. By leveraging reference latents and conditioning inputs, the node modifies the model's behavior to incorporate these references, providing more control over the motion aspects of the generated content. This can be especially beneficial in animation and video generation tasks where consistent and realistic motion is crucial.
The model
parameter represents the AI model that will be modified to incorporate the reference motion data. This is a required input and should be a pre-trained model that supports motion reference integration.
The ref_latents
parameter contains the latent representations of the reference motion data. These latents are used to guide the model in generating outputs that follow the specified motion patterns. This is a required input and should be provided in the appropriate latent format.
The enabled
parameter is a boolean that determines whether the reference motion application is active. If set to True
, the node will apply the reference motion to the model. If set to False
, the model will remain unchanged. The default value is True
.
The positive
parameter is a conditioning input that provides positive examples or prompts to guide the model's behavior. This input helps the model understand the desired motion characteristics. It is a required input and should be provided in the appropriate conditioning format.
The negative
parameter is a conditioning input that provides negative examples or prompts to guide the model's behavior. This input helps the model understand what to avoid in the generated motion. It is a required input and should be provided in the appropriate conditioning format.
The start_percent
parameter is a float that specifies the starting point of the reference motion application as a percentage of the total motion duration. It ranges from 0.0 to 1.0, with a default value of 0.0. This parameter allows you to control when the reference motion starts influencing the model.
The end_percent
parameter is a float that specifies the ending point of the reference motion application as a percentage of the total motion duration. It ranges from 0.0 to 1.0, with a default value of 0.70. This parameter allows you to control when the reference motion stops influencing the model.
The ref_settings
parameter contains the settings for the reference motion application. These settings define how the reference motion data is applied to the model. This is a required input and should be provided in the appropriate settings format.
The model
output is the modified AI model that has been updated to incorporate the reference motion data. This output allows you to use the model in subsequent nodes or processes, ensuring that the generated outputs follow the desired motion patterns.
ref_latents
parameter is correctly prepared and matches the format expected by the model to avoid errors during execution.start_percent
and end_percent
parameters to fine-tune the duration and influence of the reference motion on the model's output, optimizing the results for specific tasks.enabled
parameter to quickly toggle the reference motion application on and off, allowing you to compare the model's behavior with and without the reference motion.ref_latents
parameter are at least as long as the sample latents required by the model.positive
or negative
conditioning inputs are not in the correct format.enabled
parameter is set to False
, so the reference motion is not being applied.enabled
parameter to True
if you want to apply the reference motion to the model.© Copyright 2024 RunComfy. All Rights Reserved.