How can I achieve multiple-object detection with Amazon Rekognition Custom Labels?

For object detection, Amazon Rekognition Custom Labels can train a multi-class/multi-label object detection model with up to 250 labels per dataset. You can label more than one object with the same class in a single image during data labeling using the object drawing bbox.

Amazon Custom Labels can return up to 50 labels per image.

For information on quotas in Amazon Rekognition Custom Labels, see [Guidelines and quotas in Amazon Rekognition Custom Labels][1]. [1]: https://docs.aws.amazon.com/rekognition/latest/customlabels-dg/limits.html