AssembleNet++: Assembling Modality Representations via Attention Connections.

Michael S. Ryoo, A. J. Piergiovanni, Juhana Kangaspunta, Anelia Angelova

Browse the full ECCV paper archive.