We present SMURF, a method for unsupervised learning of optical flow tha...
General-purpose robotic systems must master a large repertoire of divers...
A common strategy to video understanding is to incorporate spatial and m...
Robots have to face challenging perceptual settings, including changes i...
One of the most challenging aspects of real-world reinforcement learning...
We systematically compare and analyze a set of key components in unsuper...
Mapping and localization, preferably from a small number of observations...
Resampling is a key component of sample-based recursive state estimation...
Estimating the 3D pose of desktop objects is crucial for applications su...
We present a novel approach to weakly supervised object detection. Inste...
We present a novel method for simultaneous learning of depth, egomotion,...
We present differentiable particle filters (DPFs): a differentiable
impl...
We propose position-velocity encoders (PVEs) which learn---without
super...
Supervised, semi-supervised, and unsupervised learning estimate a functi...