Load Video FFmpeg (Upload) 🎥🅥🅗🅢(LoadVideoFFmpegUpload)
The main function of this node is to read video files from ComfyUI's input directory and decompress them frame by frame into image tensors.
Load Video F Fmpeg Upload
The Load Video FFmpeg (Upload) 🎥🅥🅗🅢 node primarily reads video files from ComfyUI's input directory and decompresses them frame by frame into image tensors.
The Load Video FFmpeg (Upload) 🎥🅥🅗🅢 node is a core component of the Video Helper Suite toolkit, specifically designed to load user-uploaded video files into ComfyUI workflows. Built upon the powerful FFmpeg multimedia framework, it efficiently decodes and processes various common video formats, converting video content into image sequences, audio, and metadata that can be handled by subsequent nodes.
Node Functionality
The primary function of this node is to read video files from ComfyUI's input directory and decompress them frame by frame into image tensors. This is crucial for workflows that require image generation, editing, inpainting, or style transfer based on video frames. Users can precisely control the start time, frame rate, and output image dimensions, providing a flexible starting point for complex video post-processing tasks.
Node Parameter Description - Load Video FFmpeg (Upload) 🎥🅥🅗🅢
Connection Inputs (Inputs)
| Parameter Name | Data Type | Required | Default Value | Range/Options | Description |
|---|---|---|---|---|---|
video | COMBO | Yes | - | - | Select the video file to load. The file list is sourced from the ComfyUI input directory, displaying only supported video formats (e.g., .mp4, .mov, .avi). |
meta_batch | VHS_BATCHMANAGER | No | - | - | Connect a VHS_BatchManager for metadata batch processing. |
vae | VAE | No | - | - | Connect a VAE model to directly encode images into latent space representations. |
Control Parameters (Parameters)
| Parameter Name | Data Type | Required | Default Value | Range/Options | Description |
|---|---|---|---|---|---|
force_rate | COMBO | Yes | 0 | 0 - 60 (Step: 1) | Force a specific output frame rate for images, range 0-60, step 1. Uses the video's original frame rate when set to 0. |
custom_width | INT | Yes | 0 | 0 - DIMMAX | Customize the output image width, range 0 - DIMMAX. Maintains the video's original width when set to 0. |
custom_height | INT | Yes | 0 | 0 - DIMMAX | Customize the output image height, range 0 - DIMMAX. Maintains the video's original height when set to 0. |
frame_load_cap | INT | Yes | 0 | 0 - BIGMAX (Step: 1) | Limit the maximum number of frames to load, range 0 - BIGMAX, step 1. Loads all frames when set to 0. |
start_time | FLOAT | Yes | 0 | 0 - BIGMAX (Step: .001) | Set the start time for video loading (unit: seconds), range 0 - BIGMAX, step 0.001. |
Output
| Parameter Name | Data Type | Description |
|---|---|---|
| IMAGE | UNKNOWN | The output video frame sequence, formatted as image tensors. |
| mask | MASK | The mask extracted from videos with an alpha channel. |
| audio | AUDIO | The audio data separated from the video. |
| video_info | VHS_VIDEOINFO | Video metadata including frame rate, duration, and total frame count. |
Usage Scenarios
Typical usage scenarios for this node include: generating frame-by-frame animations for videos (e.g., using Stable Diffusion for repainting), extracting audio from videos for separate processing, creating masks based on video content, or simply batch-processing video screenshots. By outputting rich video information (such as frame rate, duration, total frame count), it enables subsequent nodes to perform automated processing based on accurate video attributes.
Load Video FFmpeg (Upload) 🎥🅥🅗🅢 Node Source Code Link
The Load Video FFmpeg (Upload) 🎥🅥🅗🅢 node is from the ComfyUI-VideoHelperSuite node package.
Comments
Sign in with GitHub to join the discussion.