World Labs unveils Atlas, a single AI model that generates 3D worlds from a few photos

world labs, co-founded by ai researcher fei-fei li, has announced atlas, a world model that generates, reconstructs, and simulates 3d scenes from just a few images. atlas is trained from scratch on text, images, video, and 3d data, and uses a shared spatial understanding, called 'spatial context,' to generate each new frame or viewpoint. the model can output up to one minute of video at 1440p, and can handle over a hundred input images, making it a significant advancement in ai's ability to understand 3d space.