Dynamic Zone

Class: DynamicZonesBlockV1

Source: inference.core.workflows.core_steps.transformations.dynamic_zones.v1.DynamicZonesBlockV1

The DynamicZoneBlock is a transformer block designed to simplify polygon so it's geometrically convex and then reduce number of vertices to requested amount. This block is best suited when Zone needs to be created based on shape of detected object (i.e. basketball field, road segment, zebra crossing etc.) Input detections should be filtered and contain only desired classes of interest.

Type identifier

Use the following identifier in step "type" field: roboflow_core/dynamic_zone@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
required_number_of_vertices	int	Keep simplifying polygon until number of vertices matches this number.	✅
scale_ratio	float	Expand resulting polygon along imaginary line from centroid to edge by this ratio.	✅
apply_least_squares	bool	Apply least squares algorithm to fit resulting polygon edges to base contour.	✅
midpoint_fraction	float	Fraction of vertices to keep in the middle of each edge before fitting least squares line. This parameter is useful when vertices of convex polygon are not aligned with edge that would be otherwise fitted to points closer to the center of each edge..	✅

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Available Connections

Compatible Blocks

Check what blocks you can connect to Dynamic Zone in version v1.

• inputs: Identify Changes, Line Counter, Template Matching, Model Monitoring Inference Aggregator, Detections Transformation, Camera Focus, VLM as Detector, Perspective Correction, Line Counter, Detections Stitch, Clip Comparison, Dynamic Zone, Time in Zone, Path Deviation, Slack Notification, SIFT Comparison, Detection Offset, Detections Consensus, Detections Stabilizer, Velocity, Pixel Color Count, Gaze Detection, Bounding Rectangle, Detections Filter, Cosine Similarity, Time in Zone, Webhook Sink, Segment Anything 2 Model, Roboflow Dataset Upload, Identify Outliers, Roboflow Custom Metadata, Path Deviation, Distance Measurement, VLM as Classifier, JSON Parser, Local File Sink, Twilio SMS Notification, Email Notification, PTZ Tracking (ONVIF).md), Dynamic Crop, Detections Classes Replacement, Image Contours, Instance Segmentation Model, SIFT Comparison, Instance Segmentation Model, VLM as Detector, Roboflow Dataset Upload, VLM as Classifier
• outputs: Florence-2 Model, Model Monitoring Inference Aggregator, Label Visualization, Florence-2 Model, Triangle Visualization, Model Comparison Visualization, Cache Set, Detections Transformation, Line Counter Visualization, Circle Visualization, Detections Stitch, Trace Visualization, Byte Tracker, Multi-Label Classification Model, Object Detection Model, Path Deviation, Clip Comparison, Detections Consensus, Velocity, Detection Offset, Gaze Detection, Reference Path Visualization, Llama 3.2 Vision, Bounding Rectangle, Detections Filter, Polygon Visualization, Roboflow Dataset Upload, Time in Zone, Segment Anything 2 Model, Detections Merge, Roboflow Custom Metadata, Single-Label Classification Model, Path Deviation, VLM as Classifier, Keypoint Visualization, Crop Visualization, Color Visualization, Ellipse Visualization, Google Gemini, Dynamic Crop, OpenAI, Instance Segmentation Model, Multi-Label Classification Model, Dot Visualization, Instance Segmentation Model, Roboflow Dataset Upload, Keypoint Detection Model, Keypoint Detection Model, Stability AI Inpainting, Line Counter, Single-Label Classification Model, Template Matching, Corner Visualization, Overlap Filter, Background Color Visualization, Polygon Zone Visualization, Byte Tracker, VLM as Detector, Grid Visualization, Perspective Correction, Line Counter, Clip Comparison, Dynamic Zone, Blur Visualization, Classification Label Visualization, Time in Zone, Slack Notification, SIFT Comparison, Byte Tracker, OpenAI, Detections Stabilizer, YOLO-World Model, Anthropic Claude, Size Measurement, Webhook Sink, Buffer, Mask Visualization, Bounding Box Visualization, Distance Measurement, Pixelate Visualization, Twilio SMS Notification, Email Notification, PTZ Tracking (ONVIF).md), Object Detection Model, Detections Classes Replacement, Halo Visualization, LMM For Classification, VLM as Detector, VLM as Classifier

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Dynamic Zone in version v1 has.

Bindings

• input

  • predictions (instance_segmentation_prediction): .
  • required_number_of_vertices (integer): Keep simplifying polygon until number of vertices matches this number.
  • scale_ratio (float): Expand resulting polygon along imaginary line from centroid to edge by this ratio.
  • apply_least_squares (boolean): Apply least squares algorithm to fit resulting polygon edges to base contour.
  • midpoint_fraction (float_zero_to_one): Fraction of vertices to keep in the middle of each edge before fitting least squares line. This parameter is useful when vertices of convex polygon are not aligned with edge that would be otherwise fitted to points closer to the center of each edge..

• output

  • zones (list_of_values): List of values of any type.
  • predictions (instance_segmentation_prediction): Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object.
  • simplification_converged (boolean): Boolean flag.

Example JSON definition of step Dynamic Zone in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/dynamic_zone@v1",
    "predictions": "$segmentation.predictions",
    "required_number_of_vertices": 4,
    "scale_ratio": 1.05,
    "apply_least_squares": true,
    "midpoint_fraction": 0.9
}