We present a subset selection algorithm designed to work with arbitrary ...
Mixup is a regularization technique that artificially produces new sampl...
Neural networks tend to achieve better accuracy with training if they ar...
Single image pose estimation is a fundamental problem in many vision and...
It is generally believed that robust training of extremely large network...
Deep learning has yielded extraordinary results in vision and natural
la...
Representational learning hinges on the task of unraveling the set of
un...
We can compress a neural network while exactly preserving its underlying...
We propose a novel technique to register sparse 3D scans in the absence ...
Deep neural networks have been successful in many predictive modeling ta...
Popular 3D scan registration projects, such as Stanford digital Michelan...
Two networks are equivalent if they produce the same output for any give...
3D scan registration is a classical, yet a highly useful problem in the
...
We present 3DRegNet, a deep learning algorithm for the registration of 3...
This paper proposes a novel and exact method to reconstruct line-based 3...
One form of characterizing the expressiveness of a piecewise linear neur...
Edge detection is among the most fundamental vision problems for its rol...
We propose a minimal solution for pose estimation using both points and ...
In this paper, we propose VLASE, a framework to use semantic edge featur...
The holy grail of deep learning is to come up with an automatic method t...
This paper considers a different aspect of anatomical face modeling:
kin...
Vanishing points and vanishing lines are classical geometrical concepts ...
A family of super deep networks, referred to as residual networks or Res...
In recent years, it is common practice to extract fully-connected layer ...
In this paper, we study the representational power of deep neural networ...
We propose a novel approach for group elevator scheduling by formulating...
Boundary and edge cues are highly beneficial in improving a wide variety...
We propose a layered street view model to encode both depth and semantic...
Submodular function minimization is a key problem in a wide variety of
a...