Search papers, labs, and topics across Lattice.
4
0
5
0
TAO transforms the way we handle overconfident visual recognition failures, ensuring robust performance in real-time robotic applications.
SLAM achieves a breakthrough in Visual Place Recognition by ensuring robust performance in new environments without sacrificing accuracy or requiring complex architectural changes.
Even with 40% label noise, FlatManifold maintains robust performance by leveraging intrinsic manifold properties to counteract gradient corruption.
Flattening feature manifolds can drastically enhance localization accuracy in visual place recognition, even with sparse anchor data.
