Search papers, labs, and topics across Lattice.
4
1
5
Predicting unexplored areas in 3D indoor scenes could revolutionize robotic exploration and object search capabilities.
Language observations can be transformed into precise spatial distributions, enhancing robot navigation in complex environments.
Imagine a robot that can build a 3D map granular enough to manipulate a stove knob or navigate a whole apartment, adapting on the fly to whatever task you throw at it.
Robots can now "see" hidden objects and understand articulation by learning from human egocentric video, even if they can't physically explore those areas themselves.
