Search before asking

• I have searched the Autodistill issues and found no similar feature requests.

Question

On the docs, it says we can use images as ontologies for labeling: "Other Ontologies may, for instance, use a CLIP vector or example images instead of a text caption".

However, I haven't found a single implementation of this, nor any documentation on how to do so. I have some proprietary objects I would like to detect and label, which would be difficult to achieve using CaptionOntology.

Additional

No response