ComfyUIDeployExternalVideo Node Documentation

Overview

The ComfyUIDeployExternalVideo node is part of the ComfyUI Deploy platform, specifically designed to integrate external video content into workflows. This node is geared towards handling video inputs, enabling content creators to import and process video files within ComfyUI workflows. It supports video retrieval, conversion, and manipulation, making it a versatile tool for projects that involve video content.

Functionality

This node allows users to import video files either from local directories or directly from online sources. Once imported, it provides options to process the video according to user-defined parameters, such as frame rate adjustment and resolution scaling. The node also supports a variety of video formats, ensuring compatibility with a wide range of video content.

Inputs

The node accepts the following input parameters:

• input_id: A string identifier for the input video. Can be a video filename from the local input directory or a URL pointing to an online video.
• force_rate: An integer specifying the desired frame rate for the video. Setting this to 0 retains the original frame rate.
• force_size: A selection that allows you to resize the video. Options include predefined dimensions, custom width, custom height, or no resizing.
• custom_width and custom_height: Integers specifying the dimensions to resize the video if "Custom" options are selected in force_size.
• frame_load_cap: An integer defining the maximum number of frames to load from the video.
• skip_first_frames: An integer indicating how many initial frames to skip before processing begins.
• select_every_nth: An integer that determines the periodic selection of frames from the video. For example, a value of 2 would select every second frame.
• meta_batch: An optional VHS Batch Manager input for batch processing.
• vae: An optional Variational Autoencoder input for converting video frames to latent space.
• default_video: A selectable list of video files available in the local input directory.
• display_name: An optional display name for the node in the workflow.
• description: An optional description for additional context or notes.

Outputs

The ComfyUIDeployExternalVideo node provides a variety of outputs for further use in the workflow:

• IMAGE: The processed video frames in image format.
• frame_count: An integer indicating the total number of frames processed.
• audio: Extracted audio from the video.
• video_info: Metadata about the video, including original and processed frame rates, dimensions, and duration.
• LATENT: If a Variational Autoencoder is used, this contains the video frames in latent space format.

Use Cases in ComfyUI Workflows

This node is ideal for workflows that require:

1. Video Processing: Integrate and process video data as part of a larger generative workflow. The node facilitates frame extraction and analysis.
2. Content Variation: Use frame skipping and selection options to generate variations of video content, useful in creative applications or testing scenarios.
3. Batch Operations: Utilize the meta_batch input for batch processing jobs, allowing repeated tasks across multiple videos or frames.
4. Preprocessing for AI Models: Convert videos for use with machine learning or AI models that require specific input formats, such as latent space representations.
5. Remote Video Processing: Utilize the URL input capability to process videos stored online, enabling distributed workflows.

Special Features and Considerations

• The node supports downloading and processing videos from URLs, in addition to local files, which adds great flexibility for accessing remote content.
• Multi-step frame selection settings allow for refined control over which frames are processed or analyzed.
• The node can be integrated with ComfyUI's other video helper tools and plugins, creating a comprehensive video processing ecosystem.
• It is designed to work seamlessly with ComfyUI's batch management, which is essential for workflows involving large batches of data or iterative testing.
• Consider memory and processing constraints, especially when working with high-resolution videos or complex workflows that involve multiple video files.