Detections Merge

Class: DetectionsMergeBlockV1

Source: inference.core.workflows.core_steps.transformations.detections_merge.v1.DetectionsMergeBlockV1

The DetectionsMerge block combines multiple detections into a single bounding box that encompasses all input detections. This is useful when you want to: - Merge overlapping or nearby detections of the same object - Create a single region that contains multiple detected objects - Simplify multiple detections into one larger detection

The resulting detection will have: - A bounding box that contains all input detections - The classname of the merged detection is set to "merged_detection" by default, but can be customized via the class_name parameter - The confidence is set to the lowest confidence among all detections

Type identifier

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

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
class_name	str	The class name to assign to the merged detection..	❌

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 Detections Merge in version v1.

• inputs: Google Vision OCR, Keypoint Detection Model, Time in Zone, Detections Filter, YOLO-World Model, PTZ Tracking (ONVIF).md), Detection Offset, Detections Classes Replacement, Detections Transformation, SAM 3, Template Matching, Seg Preview, Byte Tracker, Overlap Filter, SAM 3, Object Detection Model, Path Deviation, VLM as Detector, EasyOCR, Detections Combine, Dynamic Zone, Time in Zone, Object Detection Model, Detections Stitch, Line Counter, Velocity, Moondream2, Byte Tracker, OCR Model, Instance Segmentation Model, Path Deviation, Time in Zone, Dynamic Crop, Gaze Detection, Detections Consensus, Detections Stabilizer, Instance Segmentation Model, Perspective Correction, Detections Merge, VLM as Detector, SAM 3, Byte Tracker, Keypoint Detection Model, Bounding Rectangle, Segment Anything 2 Model
• outputs: Label Visualization, Time in Zone, Line Counter, Blur Visualization, Background Color Visualization, Bounding Box Visualization, Detections Filter, PTZ Tracking (ONVIF).md), Detection Offset, Pixelate Visualization, Detections Classes Replacement, Icon Visualization, Detections Transformation, Triangle Visualization, Roboflow Dataset Upload, Model Comparison Visualization, Byte Tracker, Overlap Filter, Distance Measurement, Corner Visualization, Florence-2 Model, Color Visualization, Path Deviation, Detections Combine, Size Measurement, Circle Visualization, Time in Zone, Dot Visualization, Detections Stitch, Line Counter, Ellipse Visualization, Velocity, Model Monitoring Inference Aggregator, Byte Tracker, Path Deviation, Time in Zone, Roboflow Dataset Upload, Dynamic Crop, Detections Stabilizer, Detections Consensus, Crop Visualization, Detections Merge, Perspective Correction, Florence-2 Model, Roboflow Custom Metadata, Trace Visualization, Byte Tracker, Stitch OCR Detections, Segment Anything 2 Model

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Detections Merge in version v1 has.

Bindings

• input

  • predictions (Union[instance_segmentation_prediction, object_detection_prediction, keypoint_detection_prediction]): Object detection predictions to merge into a single bounding box..

• output

  • predictions (object_detection_prediction): Prediction with detected bounding boxes in form of sv.Detections(...) object.

Example JSON definition of step Detections Merge in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/detections_merge@v1",
    "predictions": "$steps.object_detection_model.predictions",
    "class_name": "<block_does_not_provide_example>"
}