I recall this same solution being posted in the late 90’s or early 2000’s. But without the machine learning. Or to be more specific it was listed as image processing / Shape detection and I don’t recall the exact technology used to achieve this at the time. The camera set up was modified to show the cat as a silhouette against a blown out white background reducing the problem from image processing a photo to image processing / comparing the shape of the silhouette with a DB of previously tagged silhouettes.
2 Likes
