Learning what is where from unlabeled images: joint localization and clustering of foreground objects

Ashok Chandrashekar, Lorenzo Torresani, Richard Granger