Google Unveils Lumiere, a New AI Model for Video Generation

Lumiere Employs Novel 'Space-Time U-Net Architecture' for Realistic Video Clips

Google has unveiled Lumiere, a new AI model capable of generating brief yet realistic video clips.
Lumiere employs a novel 'Space-Time U-Net architecture' that generates the entire video clip at once, through a single pass in the AI model.
Lumiere's capabilities include video stylization, Cinemagraphs, and Video Inpainting feature.
Lumiere can generate 'low-resolution,' 1024×1024 videos lasting no more than 5 seconds.
Lumiere was trained on a dataset containing 30 million videos along with their text descriptions.