Object Detection & Recognition

Objects are located and labelled in an image, from general-purpose classes out of the box to a class list trained on your own material where the general set is too coarse. Each instance returns a box, a label and a confidence value. Detection can be limited to the classes you care about, which cuts both cost and false positives.

You can use the models we operate or bring your own model and we run it on your cluster. Work arrives as an API call, a batch over a folder or bucket, or a resident on-premise service, and results return as structured output with a per-job audit record.