Image Maximum Node Documentation

Overview

The Image Maximum Node is a component within the VideoX-Fun repository designed for use with ComfyUI, an interface for video and image transformation. The primary function of this node is to compute the pixel-wise maximum of two given video inputs. This process involves comparing corresponding pixels across the frames of two videos and retaining the maximum value from each comparison to produce a new video output.

Inputs

Video Inputs

1. video_1: The first video input. This video is a collection of frames, where each frame is a two-dimensional array of pixels with color channel values. The dimensions of the video include its length (number of frames), height, width, and the number of color channels (typically three, for RGB).

2. video_2: The second video input, similar in structure to video_1. It should ideally have the same resolution (height and width) and number of frames as video_1 to avoid any rescaling. However, if they differ in resolution, the second video will be resized to match the first before processing.

Outputs

Video Output

• image: The output is a video where each pixel value is the maximum of the corresponding pixels from the input videos. This results in a video that effectively combines the most prominent pixel values from both inputs, often resulting in a more vibrant and detailed visual representation.

Usage in ComfyUI Workflows

The Image Maximum Node can be used in video processing workflows where the goal is to enhance or combine features from two separate video inputs. It can be particularly useful in scenarios where you want to merge two videos to emphasize certain aspects, such as contrast or color intensity.

Example Workflow

1. Input Preparation: Start by importing two video files into the ComfyUI environment. Ensure that both videos are pre-processed to align in terms of frame count and resolution.

2. Node Integration: Place the Image Maximum Node within the workflow. Connect the node's inputs to the respective video sources (video_1 and video_2).

3. Video Processing: The node will process the input videos and output a new video file with enhanced visual qualities derived from the pixel-wise maximum operation.

4. Output Utilization: The resulting video can be utilized for further editing, effects application, or directly exported as a final product.

Special Features and Considerations

• Resizing Capability: If the input videos have differing resolutions, video_2 will be automatically resized to match the dimensions of video_1. This ensures consistent pixel mapping during the maximum value computation.

• Efficiency: The node operates efficiently by leveraging optimized operations that handle the manipulation of potentially large video files within the ComfyUI environment.

• Compatibility: The Image Maximum Node is designed to integrate seamlessly into ComfyUI workflows, offering users flexibility in video processing tasks without requiring in-depth technical knowledge of the underlying operations.

Overall, the Image Maximum Node is a powerful tool for users looking to enhance or creatively combine video content within the ComfyUI framework, providing a straightforward yet effective method for image maximization tasks.