Learning Aligned Cross-Modal Representations from Weakly Aligned Data.

Llus Castrejn, Yusuf Aytar, Carl Vondrick, Hamed Pirsiavash, Antonio Torralba

Browse the full CVPR paper archive.