Deep neural networks (NNs) are known to lack uncertainty estimates and
s...
Approximate inference in Gaussian process (GP) models with non-conjugate...
Simultaneous localization and mapping (SLAM) is the task of building a m...
Sequential learning with Gaussian processes (GPs) is challenging when ac...
Diffusion processes are a class of stochastic differential equations (SD...
Dynamic neural networks are a recent technique that promises a remedy fo...
The dynamic Schrödinger bridge problem provides an appealing setting for...
Mixup is a popular data augmentation technique for training deep neural
...
Gaussian process training decomposes into inference of the (approximate)...
Gaussian processes (GPs) are the main surrogate functions used for seque...
We introduce visual hints expansion for guiding stereo matching to impro...
Source-free domain adaptation (SFDA) aims to adapt a classifier to an
un...
While diffusion models have shown great success in image generation, the...
It is prevalent and well-observed, but poorly understood, that two machi...
The fusion of camera sensor and inertial data is a leading method for
eg...
Gaussian processes (GPs) provide a principled and direct approach for
in...
Sparse variational Gaussian process (SVGP) methods are a common choice f...
We introduce a scalable approach to Gaussian process inference that comb...
We formulate natural gradient variational inference (VI), expectation
pr...
Simulation-based techniques such as variants of stochastic Runge-Kutta a...
Neural network models are known to reinforce hidden data biases, making ...
We present HybVIO, a novel hybrid approach for combining filtering-based...
Gaussian processes (GPs) are important probabilistic tools for inference...
Approximate Bayesian inference methods that scale to very large datasets...
We introduce a principled approach for synthesizing new views of a scene...
Current robot platforms available for research are either very expensive...
We introduce a new family of non-linear neural network activation functi...
We propose a method for fusing stereo disparity estimation with
movement...
We formulate approximate Bayesian inference in non-conjugate temporal an...
Gaussian process (GP) regression with 1D inputs can often be performed i...
Robust and accurate six degree-of-freedom tracking on portable devices
r...
We propose Deep Residual Mixture Models (DRMMs) which share the many
des...
Gaussian processes are powerful non-parametric probabilistic models for
...
We introduce a novel autoencoder model that deviates from traditional
au...
We derive a principled framework for encoding prior knowledge of informa...
Multi-output Gaussian processes (MOGPs) leverage the flexibility and
int...
Modern smartphones have all the sensing capabilities required for accura...
We propose a novel idea for depth estimation from unstructured multi-vie...
We build on recent advances in progressively growing generative autoenco...
Gaussian processes (GPs) provide a powerful framework for extrapolation,...
This paper presents a novel method, MaskMVS, to solve depth estimation f...
A typical audio signal processing pipeline includes multiple disjoint
an...
Gaussian processes provide a flexible framework for forecasting, removin...
In audio signal processing, probabilistic time-frequency models have man...
Strapdown inertial navigation systems are sensitive to the quality of th...
The lack of realistic and open benchmarking datasets for pedestrian
visu...
We introduce a novel generative autoencoder network model that learns to...
Camera calibration for estimating the intrinsic parameters and lens
dist...
We present a method for scalable and fully 3D magnetic field simultaneou...
This paper addresses the modeling and simulation of progressive changes ...