NormalizeVideoLatentStart

conditioning/video_models

NormalizeVideoLatentStart

Normalizes the initial frames of a video latent to match the mean and standard deviation of subsequent reference frames. Helps reduce differences between the starting frames and the rest of the video.

Example

JSON Example

{
  "class_type": "NormalizeVideoLatentStart",
  "inputs": {
    "latent": [
      "node_id",
      0
    ],
    "start_frame_count": 4,
    "reference_frame_count": 5
  }
}

This example shows required inputs only. Connection values like ["node_id", 0] should reference actual node IDs from your workflow.

Inputs

Name	Type	Status	Constraints	Default
latent	LATENT	required	-	-
start_frame_count?	INT	required	min: 1, max: 16384, step: 1	4
reference_frame_count?	INT	required	min: 1, max: 16384, step: 1	5

Outputs

Index	Name	Type	Is List	Connection Reference
0	latent	LATENT	No	["{node_id}", 0]

How to connect to these outputs

To connect another node's input to an output from this node, use the connection reference format:

["node_id", output_index]

Where node_id is the ID of this NormalizeVideoLatentStart node in your workflow, and output_index is the index from the table above.

Example

If this node has ID "5" in your workflow:

• latent (LATENT): ["5", 0]

Package: comfyui-extras