Detections Stabilizer

Class: StabilizeTrackedDetectionsBlockV1

Source: inference.core.workflows.core_steps.transformations.stabilize_detections.v1.StabilizeTrackedDetectionsBlockV1

This block stores last known position for each bounding box If box disappears then this block will bring it back so short gaps are filled with last known box position The block requires detections to be tracked (i.e. each object must have unique tracker_id assigned, which persists between frames) WARNING: this block will produce many short-lived bounding boxes for unstable trackers!

Type identifier

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

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
smoothing_window_size	int	Predicted movement of detection will be smoothed based on historical measurements of velocity, this parameter controls number of historical measurements taken under account when calculating smoothed velocity. Detections will be removed from generating smoothed predictions if they had been missing for longer than this number of frames..	✅
bbox_smoothing_coefficient	float	Bounding box smoothing coefficient applied when given tracker_id is present on current frame. This parameter must be initialized with value between 0 and 1.	✅

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 Stabilizer in version v1.

• inputs: Bounding Box Visualization, Stability AI Outpainting, Detections Merge, Moondream2, Overlap Filter, Image Threshold, Detections Stabilizer, Model Comparison Visualization, SIFT Comparison, Image Slicer, Corner Visualization, Background Color Visualization, Distance Measurement, Image Contours, Time in Zone, Mask Visualization, QR Code Generator, Classification Label Visualization, Detections Transformation, Trace Visualization, Polygon Visualization, Perspective Correction, Instance Segmentation Model, Path Deviation, Grid Visualization, SIFT Comparison, Clip Comparison, Instance Segmentation Model, Byte Tracker, Image Convert Grayscale, PTZ Tracking (ONVIF).md), Line Counter, Dot Visualization, Relative Static Crop, Ellipse Visualization, Object Detection Model, Pixel Color Count, Dynamic Zone, Halo Visualization, Polygon Zone Visualization, Time in Zone, VLM as Detector, Icon Visualization, Triangle Visualization, Crop Visualization, Pixelate Visualization, Path Deviation, Stitch Images, SIFT, Color Visualization, Blur Visualization, Camera Focus, Absolute Static Crop, Label Visualization, Line Counter Visualization, Detection Offset, Detections Consensus, Reference Path Visualization, Velocity, Camera Calibration, Image Blur, Dynamic Crop, Byte Tracker, Segment Anything 2 Model, Circle Visualization, Template Matching, YOLO-World Model, Identify Outliers, Byte Tracker, Image Slicer, Depth Estimation, Bounding Rectangle, Image Preprocessing, Stability AI Inpainting, Line Counter, Keypoint Visualization, Identify Changes, Detections Classes Replacement, Object Detection Model, Stability AI Image Generation, VLM as Detector, Detections Filter, Detections Stitch, Google Vision OCR
• outputs: Model Monitoring Inference Aggregator, Bounding Box Visualization, Stitch OCR Detections, Color Visualization, Detections Merge, Blur Visualization, Overlap Filter, Detections Stabilizer, Model Comparison Visualization, Corner Visualization, Distance Measurement, Background Color Visualization, Time in Zone, Florence-2 Model, Label Visualization, Mask Visualization, Detection Offset, Detections Consensus, Detections Transformation, Trace Visualization, Polygon Visualization, Velocity, Perspective Correction, Florence-2 Model, Path Deviation, Dynamic Crop, Byte Tracker, Roboflow Dataset Upload, Circle Visualization, Segment Anything 2 Model, Byte Tracker, Byte Tracker, PTZ Tracking (ONVIF).md), Line Counter, Bounding Rectangle, Dot Visualization, Ellipse Visualization, Stability AI Inpainting, Line Counter, Dynamic Zone, Time in Zone, Halo Visualization, Roboflow Custom Metadata, Detections Classes Replacement, Icon Visualization, Triangle Visualization, Roboflow Dataset Upload, Crop Visualization, Size Measurement, Detections Filter, Detections Stitch, Pixelate Visualization, Path Deviation

Input and Output Bindings

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

Bindings

• input

  • image (image): not available.
  • detections (Union[object_detection_prediction, instance_segmentation_prediction]): Tracked detections.
  • smoothing_window_size (integer): Predicted movement of detection will be smoothed based on historical measurements of velocity, this parameter controls number of historical measurements taken under account when calculating smoothed velocity. Detections will be removed from generating smoothed predictions if they had been missing for longer than this number of frames..
  • bbox_smoothing_coefficient (float_zero_to_one): Bounding box smoothing coefficient applied when given tracker_id is present on current frame. This parameter must be initialized with value between 0 and 1.

• output

  • tracked_detections (Union[object_detection_prediction, instance_segmentation_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.

Example JSON definition of step Detections Stabilizer in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/stabilize_detections@v1",
    "image": "<block_does_not_provide_example>",
    "detections": "$steps.object_detection_model.predictions",
    "smoothing_window_size": 5,
    "bbox_smoothing_coefficient": 0.2
}