Suppose I take a photo of a tomato, a piece of bread, and cheese on a table. Is there a feature that crops each ingredient into separate photos or at least adds some kind of numerical or string label to it?
Hey @arun_ajay! Unfortunately our Food Model can only return labels and corresponding probabilities of those labels appearing in the photo somewhere, but not any localized bounding box information. It's definitely something that we're working on though. Can I ask what your use case is with the localization?
