4M-21: An Any-to-Any Vision Model for Tens of Tasks and Modalities
Roman Bachmann, Afshin Dehghan, Mingfei Gao, Ali Garjani, David Griffiths, Jiaming Hu, Oğuzhan Fatih Kar, David Mizrahi, Amir Zamir
Topics & Concepts
Computer scienceModalitiesComputer visionArtificial intelligenceHuman–computer interactionModality (human–computer interaction)Focus (optics)PsychologyKey (lock)Feature (linguistics)Visual perception and processing mechanismsEmbodied and Extended CognitionFace Recognition and Perception