AVFormer: Injecting Vision into Frozen Speech Models for Zero-Shot AV-ASR.

Paul Hongsuck Seo, Arsha Nagrani, Cordelia Schmid

Browse the full CVPR paper archive.