Search papers, labs, and topics across Lattice.
5
0
5
Learning robot skills from human videos offers a scalable path to generalist robotic systems, but faces challenges in bridging the gap between human demonstrations and robot execution.
Forget rigid transforms: CAL2M's elastic sub-map alignment unlocks globally consistent kilometer-scale SLAM with Visual Geometry Foundation Models.
LLMs can't hold a candle to a well-tuned POMDP solver when it comes to efficiently finding objects in the real world.
Achieve photorealistic multi-traversal 3D reconstruction by disentangling illumination from material properties, even when the same scene is captured under drastically different lighting conditions.
Differentiable collision checking in configuration space, previously a major hurdle, is now achievable with zero-shot generalization thanks to CSSDF-Net.
