Microsoft Research's Mirage gives video generation a persistent spatial memory that doesn't forget what's around the corner

Microsoft Research has introduced Mirage, a video generation model that maintains spatial consistency by storing scene data in latent space rather than pixel-based point clouds. By avoiding resource-intensive calculations, this approach allows for long, coherent camera movements while significantly reducing memory and processing requirements. This development provides a more efficient method for generating navigable virtual environments, addressing a recurring challenge in maintaining logical structures during complex video rendering.