San Francisco-based World Labs has announced Atlas, an AI model that can perform tasks such as world generation, reconstruction and simulation:
- Camera-Controlled Generation: generate images and videos from one or more images with camera control, outputting up to 1 minute of video at 1440p.
- Spatial Reconstruction: reconstruct real world scenes from one to dozens of input images. Atlas generates both image frames from novel views and explicit 3D outputs.
- Space-Time Simulation: model space and time from input videos, reframing videos for visual effects and enabling Real-to-Sim workflows for robotics.
- Image Generation: generate images and 360 panoramas from text; it can follow complex prompts and generate a wide variety of visual styles.
Watch a video and find out more on World Labs’ site.
