Detection Offset

Class: DetectionOffsetBlockV1

Source: inference.core.workflows.core_steps.transformations.detection_offset.v1.DetectionOffsetBlockV1

Apply a fixed offset to the width and height of a detection.

You can use this block to add padding around the result of a detection. This is useful to ensure that you can analyze bounding boxes that may be within the region of an object instead of being around an object.

Type identifier

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

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
offset_width	int	Offset for box width..	✅
offset_height	int	Offset for box height..	✅
units	str	Units for offset dimensions..	❌

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 Detection Offset in version v1.

• inputs: Detection Offset, Moondream2, Path Deviation, VLM as Detector, Distance Measurement, Detections Classes Replacement, Gaze Detection, Bounding Rectangle, Image Contours, Detections Stitch, Template Matching, Instance Segmentation Model, Byte Tracker, Keypoint Detection Model, SIFT Comparison, Object Detection Model, Path Deviation, Line Counter, Detections Stabilizer, Detections Transformation, Byte Tracker, Google Vision OCR, Dynamic Zone, Perspective Correction, Object Detection Model, Detections Filter, SIFT Comparison, Instance Segmentation Model, Pixel Color Count, YOLO-World Model, VLM as Detector, Detections Merge, Velocity, Segment Anything 2 Model, Time in Zone, Time in Zone, Line Counter, Keypoint Detection Model, Overlap Filter, Byte Tracker, Dynamic Crop, Detections Consensus
• outputs: Size Measurement, Detection Offset, Roboflow Custom Metadata, Path Deviation, Florence-2 Model, Distance Measurement, Label Visualization, Detections Classes Replacement, Background Color Visualization, Bounding Rectangle, Detections Stitch, Byte Tracker, Triangle Visualization, Path Deviation, Line Counter, Detections Stabilizer, Detections Transformation, Byte Tracker, Keypoint Visualization, Stitch OCR Detections, Roboflow Dataset Upload, Dynamic Zone, Bounding Box Visualization, Perspective Correction, Halo Visualization, Circle Visualization, Ellipse Visualization, Crop Visualization, Color Visualization, Dot Visualization, Pixelate Visualization, Detections Filter, Model Comparison Visualization, Detections Merge, Velocity, Roboflow Dataset Upload, Segment Anything 2 Model, Polygon Visualization, Time in Zone, Line Counter, Time in Zone, Trace Visualization, Corner Visualization, Blur Visualization, Model Monitoring Inference Aggregator, Stability AI Inpainting, Mask Visualization, Florence-2 Model, Overlap Filter, Byte Tracker, Dynamic Crop, Detections Consensus

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Detection Offset in version v1 has.

Bindings

• input

  • predictions (Union[instance_segmentation_prediction, object_detection_prediction, keypoint_detection_prediction]): Model predictions to offset dimensions for..
  • offset_width (integer): Offset for box width..
  • offset_height (integer): Offset for box height..

• output

  • predictions (Union[object_detection_prediction, instance_segmentation_prediction, keypoint_detection_prediction]): Prediction with detected bounding boxes in form of sv.Detections(...) object if object_detection_prediction or Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object if instance_segmentation_prediction or Prediction with detected bounding boxes and detected keypoints in form of sv.Detections(...) object if keypoint_detection_prediction.

Example JSON definition of step Detection Offset in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/detection_offset@v1",
    "predictions": "$steps.object_detection_model.predictions",
    "offset_width": 10,
    "offset_height": 10,
    "units": "Pixels"
}