WanCameraImageToVideo - ComfyUI Built-in Node Documentation

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

The WanCameraImageToVideo node converts images to video sequences by generating latent representations for video generation. It processes conditioning inputs and optional starting images to create video latents that can be used with video models. The node supports camera conditions and clip vision outputs for enhanced video generation control.

Inputs

Parameter	Data Type	Required	Range	Description
`positive`	CONDITIONING	Yes	-	Positive conditioning prompts for video generation
`negative`	CONDITIONING	Yes	-	Negative conditioning prompts to avoid in video generation
`vae`	VAE	Yes	-	VAE model for encoding images to latent space
`width`	INT	Yes	16 to MAX_RESOLUTION	Output video width in pixels (default: 832, step: 16)
`height`	INT	Yes	16 to MAX_RESOLUTION	Output video height in pixels (default: 480, step: 16)
`length`	INT	Yes	1 to MAX_RESOLUTION	Number of frames in the video sequence (default: 81, step: 4)
`batch_size`	INT	Yes	1 to 4096	Number of videos to generate simultaneously (default: 1)
`clip_vision_output`	CLIP_VISION_OUTPUT	No	-	Optional CLIP vision output for additional conditioning
`start_image`	IMAGE	No	-	Optional starting image to initialize the video sequence
`camera_conditions`	WAN_CAMERA_EMBEDDING	No	-	Optional camera embedding conditions for video generation

Note: When start_image is provided, the node uses it to initialize the video sequence and applies masking to blend the starting frames with generated content. The camera_conditions and clip_vision_output parameters are optional but when provided, they modify the conditioning for both positive and negative prompts.

Outputs

Output Name	Data Type	Description
`positive`	CONDITIONING	Modified positive conditioning with applied camera conditions and clip vision outputs
`negative`	CONDITIONING	Modified negative conditioning with applied camera conditions and clip vision outputs
`latent`	LATENT	Generated video latent representation for use with video models

WanAnimateToVideo - ComfyUI Built-in Node Documentation

WanFirstLastFrameToVideo - ComfyUI Built-in Node Documentation

​Inputs

​Outputs

Inputs

Outputs