Google has introduced a new video GenAI model named Lumiere that employs a diffusion approach called Space-Time-U-Net (STUNet) to create the video in one process instead of putting smaller still frames together.
Lumiere begins by forming a base frame from the user's prompt and proceeds to map the movement and transformation of objects within the frame to craft additional frames. It uses STUNet to focus on the actual movement based on where the created content should be at a specific time in the video, providing a competitive edge.
Its key features are image-to-video generation, stylized generation, the creation of cinemagraphs, and in-painting functionality to change colors or patterns in a portion of the video.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.