Description
Duplicate and near-duplicate images can be found with hashing and machine-learning approaches. Developers and data teams use imagededup to clean image collections and catalogs. Image datasets may contain private people or places, so results and thumbnails need privacy review.