Neural networks are known toproduce poor uncertainty estimations, and avariety of approaches have been proposed to remedy this issue.

Multi-scene absolute pose regression addresses the demand for fast and memory-efficient camera pose estimation across various real-world environments.

A general solution of the kernel space constraint is given for arbitrary representations of the Euclidean groupE(2) and its subgroups.

The embedding of such a manifold into the high-dimensional space of the data is achieved via learnable invertible transformations.

Neural Information Processing Systems http://nips.cc/