Visit ComfyUI Online for ready-to-use ComfyUI environment
Detect and track blobs in images or video frames using OpenCV's SimpleBlobDetector for movement analysis and visualization.
The AK_BlobTrack node is designed to detect and track blobs, or distinct areas of interest, within a sequence of images or video frames. Utilizing OpenCV's SimpleBlobDetector, this node isolates movement through frame differencing and thresholding techniques. It then identifies blobs based on specified criteria such as size and brightness. The node provides a visual representation by drawing outlines around detected blobs and connecting them with lines, enhancing the understanding of movement patterns. This functionality is particularly beneficial for AI artists looking to analyze motion or changes within visual media, offering a straightforward method to highlight and track dynamic elements.
This parameter represents the input image or frame sequence where blob detection will be performed. It is essential for the node to process and identify areas of interest.
Determines the number of frames to cache for frame differencing, which is crucial for detecting movement. The value ranges from 1 to 120, with a default of 1, allowing you to control the sensitivity to motion over time.
Sets the threshold for frame differencing, influencing the sensitivity to changes between frames. The range is from 0.0 to 255.0, with a default of 30.0, where higher values may ignore minor changes.
Defines the minimum threshold for blob detection, affecting the initial sensitivity to potential blobs. It ranges from 0 to 255, with a default of 50.
Specifies the maximum threshold for blob detection, setting the upper limit for blob sensitivity. The range is from 1 to 255, with a default of 220.
Controls the step size between thresholds during blob detection, impacting the granularity of detection. It ranges from 1 to 50, with a default of 10.
A boolean option to filter blobs based on their area. When set to "true" (default), only blobs within the specified area range are detected.
Sets the minimum area for blob detection, ensuring that only sufficiently large blobs are considered. The range is from 0.0 to 1e6, with a default of 25.0.
Defines the maximum area for blob detection, preventing overly large blobs from being detected. The range is from 1.0 to 1e8, with a default of 1e5.
A boolean option to detect bright blobs. When set to "true", the node focuses on bright areas, with a default of "false".
Limits the number of blobs to be detected, allowing you to focus on the most significant ones. The range is from 1 to 100, with a default of 10.
Specifies the thickness of the outline drawn around detected blobs, enhancing their visibility. The range is from 1 to 20, with a default of 2.
Sets the color of the blob outline using a hex code, with a default of "#ff0000" (red), allowing for customization of the visual output.
Controls the transparency of the blob outline, with a range from 0.0 (fully transparent) to 1.0 (fully opaque), and a default of 1.0.
Determines the thickness of lines connecting blob centers, aiding in visualizing movement paths. The range is from 1 to 20, with a default of 2.
Sets the color of the connecting lines using a hex code, with a default of "#00ff00" (green), allowing for customization of the visual output.
Controls the transparency of the connecting lines, with a range from 0.0 (fully transparent) to 1.0 (fully opaque), and a default of 1.0.
The output image with detected blobs outlined and connected by lines, providing a visual representation of movement and areas of interest.
A binary mask image where detected blobs are filled in white, useful for further processing or analysis of the detected areas.
diff_threshold
to fine-tune sensitivity to movement; lower values detect subtle changes, while higher values focus on significant motion.filter_by_area
to exclude irrelevant blobs by setting appropriate min_area
and max_area
values, ensuring only meaningful blobs are tracked.blob_outline_color
and line_color
, are specified as 6-digit hex codes, e.g., "#ff0000".cache_frames
parameter is set to a value larger than the available frames.cache_frames
does not exceed the number of frames in your input sequence. Adjust the parameter accordingly.RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Playground, enabling artists to harness the latest AI tools to create incredible art.