ComfyUI Node: Motion Ref Settings (Custom)

Motion Ref Settings (Custom):

The MotionRefSettingsCustomNode is designed to provide advanced customization for motion reference settings in your AI art projects. This node allows you to fine-tune various parameters that control how motion references are applied, offering greater flexibility and precision in your creative process. By enabling or disabling specific attention mechanisms and adjusting fidelity settings, you can achieve more nuanced and controlled motion effects. This node is particularly useful for artists looking to experiment with different motion dynamics and achieve specific visual outcomes in their work.

Motion Ref Settings (Custom) Input Parameters:

enabled

This parameter is a boolean that determines whether the motion reference settings should be applied. If set to True, the settings will be active; if set to False, the previous settings will be retained. The default value is True.

q_bank

This boolean parameter enables or disables the query bank mode. When enabled, it allows the node to use a specific set of query vectors for motion reference. The default value is False.

k_bank

This boolean parameter enables or disables the key bank mode. When enabled, it allows the node to use a specific set of key vectors for motion reference. The default value is False.

v_bank

This boolean parameter enables or disables the value bank mode. When enabled, it allows the node to use a specific set of value vectors for motion reference. The default value is False.

norm_bank

This boolean parameter enables or disables the normalization bank mode. When enabled, it allows the node to use a specific set of normalization vectors for motion reference. The default value is False.

norm_fidelity

This float parameter controls the fidelity of the normalization process. It ranges from 0.0 to 1.0, with a default value of 0.5. Higher values result in more precise normalization.

input_1, input_2, input_3, input_4, input_5, input_6, input_7, input_8

These boolean parameters control the attention mechanisms for the input vectors. Each parameter can be set to True or False, enabling or disabling attention for the corresponding input vector. The default value for each is False.

output_1, output_2, output_3, output_4, output_5, output_6, output_7, output_8, output_9, output_10, output_11, output_12

These boolean parameters control the attention mechanisms for the output vectors. Each parameter can be set to True or False, enabling or disabling attention for the corresponding output vector. The default value for each is False.

prev_settings

This optional parameter allows you to pass in previous motion reference settings. It accepts a list of MOTION_REF_SETTINGS and is useful for chaining multiple settings together.

Motion Ref Settings (Custom) Output Parameters:

MOTION_REF_SETTINGS

The output is a list of motion reference settings that have been configured based on the input parameters. This list can be used in subsequent nodes to apply the specified motion reference settings to your model.

Motion Ref Settings (Custom) Usage Tips:

• Experiment with different combinations of q_bank, k_bank, and v_bank to see how they affect the motion dynamics in your project.
• Use the norm_fidelity parameter to fine-tune the precision of the normalization process, especially when working with high-detail motion references.
• Enable only the necessary input and output attentions to optimize performance and achieve the desired visual effect.

Motion Ref Settings (Custom) Common Errors and Solutions:

"Invalid input parameter"

• Explanation: This error occurs when one or more input parameters are not set correctly.
• Solution: Double-check all input parameters to ensure they are within the allowed ranges and of the correct type.

"Previous settings not found"

• Explanation: This error occurs when the prev_settings parameter is expected but not provided.
• Solution: Ensure that you pass a valid list of previous settings if you intend to use this parameter.

"Normalization fidelity out of range"

• Explanation: This error occurs when the norm_fidelity parameter is set outside the allowed range of 0.0 to 1.0.
• Solution: Adjust the norm_fidelity parameter to be within the specified range.