Google shows off Lumiere, a space-time diffusion model for realistic AI videos

Lumiere was trained on a dataset of 30 million videos, along with their text captions, and is capable of generating 80 frames at 16 fps. The source of this data, however, remains unclear at this stage.

ai ai video ai video editing ai video generation arts & entertainment business captions conversational ai data dataset diffusion diffusion model fps generative-ai google google lumiere image-to-video lumiere ml and deep learning nlp pika 1.0 pika labs runway science shows space stability ai stable video diffusion stage tel aviv university text text-to-video text-to-video model video generation videos