Configure General Object Detection

General object detection models are pre-trained models that can be used for detecting 80 classes of common objects. The models are trained on the COCO dataset and support either bounding box or polygon labels.
Click to configure and add a General Object Detection model to your pipeline.
To add a general object detection model to your pipeline:
  1. 1.
    Navigate to the "Pipeline Blocks" tab.
  2. 2.
    Under the Prediction section, click on the General Object Detection option to configure this processing block.
  3. 3.
    Under Model Name, select which object detection model you wish to use: Default COCO (Bounding Box) or Default COCO (Polygon). The main difference between the two options is the type of label it uses for detections.
  4. 4.
    Under Classes, select which classes (labels) you want to return detections for. By default, all 80 classes are selected.
  5. 5.
    Using the slider, select the desired Confidence Threshold. This value defaults to 60%. Only detections that meet or exceed this confidence score are returned.
Select which classes you want to return detections for.
When finished, scroll down and click "Add to Pipeline" to add the block to your pipeline. This will bring you back to the "Pipeline Blocks" tab. From there you can add additional blocks and drag and drop to rearrange around other blocks as needed.