Cooperative Learning of Audio and Video Models from Self-Supervised Synchronization