Movie Gen is Meta’s AI model designed to generate high-quality video and audio from text descriptions. It can create videos up to 16 seconds long in 1080p resolution
Trained for both text-to-image and text-to-video tasks, Movie Gen generates coloured frames to form videos, reasoning about object motion and subject interactions
Movie Gen can create videos featuring a person from a real image, maintaining their identity while performing actions based on the user’s prompt
Using video-to-audio and text-to-audio techniques, Movie Gen generates 48kHz audio with cinematic sound effects, synchronised to the video input
Movie Gen can edit both generated and real videos based on user input, making precise and creative changes to elements like background and objects