DrawInstanceDiffusionTracking

The DrawInstanceDiffusionTracking node receives an image batch and tracking data of type TRACKING as input.

The DrawInstanceDiffusionTracking node is used to visualize instance segmentation and tracking results on images. Its core function is to read tracking data generated by the CreateInstanceDiffusionTracking node and draw this data (such as object bounding boxes and category labels) onto the corresponding images, thereby generating visualized images with annotation information.

Node Function

This node receives an image batch and tracking data of type TRACKING as input. Internally, it uses Matplotlib's rainbow color map to assign colors to different tracking categories and utilizes the drawing capabilities of the PIL library to overlay the detected object bounding boxes and optional category text for each image frame onto the original images. Users can customize the drawing style, such as box line width, font, and text display.

Node Parameter Description - DrawInstanceDiffusionTracking

Connection Inputs (Inputs)

Parameter Name	Data Type	Required	Default Value	Value Range/Options	Description
`image`	IMAGE	Yes	-	-	Input image batch on which to draw the tracking data.
`tracking`	TRACKING	Yes	-	-	tracking (input parameter): Tracking data from the CreateInstanceDiffusionTracking node, containing category and bounding box information.
`font`	COMBO	Yes	-	-	Font name to use for drawing text. Select from the available fonts.

Control Parameters (Parameters)

Parameter Name	Data Type	Required	Default Value	Value Range/Options	Description
`box_line_width`	INT	Yes	2	1 - 10 (step: 1)	box_line_width (input parameter): Sets the line width for drawing bounding boxes. The value range is 1 to 10.
`draw_text`	BOOLEAN	Yes	True	-	draw_text (input parameter): A boolean switch that controls whether to draw category label text next to the bounding boxes.
`font_size`	INT	Yes	20	-	font_size (input parameter): Sets the font size for drawing text.

Output

Parameter Name	Data Type	Description
image	IMAGE	image (output parameter): Outputs the visualized image with drawn tracking boxes and labels.

Usage Scenarios

This node is very useful in workflows for video analysis or sequential image processing. For example, after processing all frames of a video and obtaining object tracking data, you can use this node to draw the tracking boxes and ID information frame by frame back onto the original video images, thereby generating a visualized video with tracking trajectories. This facilitates intuitive inspection of the accuracy and stability of the tracking algorithm.

Notes

Please note that the input tracking data must come from the CreateInstanceDiffusionTracking node, and its structure must strictly correspond temporally with the input image batch. If the frame index in the tracking data exceeds the range of the image batch, the corresponding drawing operation will be skipped.