Meta AI’s OMNIVORE: A Modality-Agnostic Single Vision Model With Cross-Modal Generalization

Web: https://syncedreview.com/2022/01/24/deepmind-podracer-tpu-based-rl-frameworks-deliver-exceptional-performance-at-low-cost-191/

A Meta AI research team presents OMNIVORE, a single vision model for various visual modalities that can perform cross-modal generalization and achieves performance at par or better than traditional modality-specific models of the same size.

ai artificial intelligence computer vision computer vision & graphics cross machine learning machine learning & data science meta ml model multimodal learning research technology transformer vision