Search
The multimodal world model Atlas, developed by Li Feifei's team, centers around 'novel view prediction'. It integrates generation and reconstruction, natively processes multimodal data, and has the potential to become one of the fundamental primitives for Artificial General Intelligence (AGI). This model can significantly reduce the amount of data required for 3D reconstruction and address blind spot issues in traditional 3D reconstruction. It holds vast application prospects in fields such as creative design, construction, and robotics. Currently, Atlas has initial dynamic processing capabilities and will further optimize dynamic effects, content editing, and human-computer interaction in the future, with the goal of building a 3D virtual world system that ordinary users can interact with intuitively.
-
C114 Communication Network -
Communication Home
