We propose learning a depth covariance function with applications to
geo...
We present vMAP, an object-level dense SLAM system using neural field
re...
Neural fields can be trained from scratch to represent the shape and
app...
General scene understanding for robotics requires flexible semantic
repr...
In this paper, we propose a novel object-level mapping system that can
s...
Precise coordinated planning enables safe and highly efficient motion wh...
There are many possibilities for how to represent the map in simultaneou...
Robots need the capability of placing objects in arbitrary, specific pos...
Robots need object-level scene understanding to manipulate objects while...
We show that a distributed network of robots or other devices which make...
Understanding the structure of multiple related tasks allows for multi-t...
Joint representation of geometry, colour and semantics using a 3D neural...
Scene graphs represent the key components of a scene in a compact and
se...
We propose a novel dense mapping framework for sparse visual SLAM system...
In this article, we present a visual introduction to Gaussian Belief
Pro...
Reflecting on the last few years, the biggest breakthroughs in deep
rein...
Despite the success of reinforcement learning methods, they have yet to ...
We present ReCo, a contrastive learning framework designed at a regional...
By estimating 3D shape and instances from a single view, we can capture
...
Semantic labelling is highly correlated with geometry and radiance
recon...
We show for the first time that a multilayer perceptron (MLP) can serve ...
Spatial memory, or the ability to remember and recall specific locations...
We describe a framework for research and evaluation in Embodied AI. Our
...
Dense image alignment from RGB-D images remains a critical issue for
rea...
Robots and other smart devices need efficient object-based scene
represe...
Graph processors such as Graphcore's Intelligence Processing Unit (IPU) ...
The ability to estimate rich geometry and camera motion from monocular
i...
Humans can naturally learn to execute a new task by seeing it performed ...
We argue the case for Gaussian Belief Propagation (GBP) as a strong
algo...
We present a challenging new benchmark and learning-environment for robo...
PyRep is a toolkit for robot learning research, built on top of the virt...
Systems which incrementally create 3D semantic maps from image sequences...
Systems which incrementally create 3D semantic maps from image sequences...
Learning with auxiliary tasks has been shown to improve the generalisati...
Much like humans, robots should have the ability to leverage knowledge f...
Sum-of-squares objective functions are very popular in computer vision
a...
We propose an online object-level SLAM system which builds a persistent ...
SLAM is becoming a key component of robotics and augmented reality (AR)
...
Visual understanding of 3D environments in real-time, at low power, is a...
We have seen much recent progress in rigid object manipulation, but
inte...
The representation of geometry in real-time 3D perception systems contin...
We discuss and predict the evolution of Simultaneous Localisation and Ma...
In this paper, we propose a novel multi-task learning architecture, whic...
In this paper we investigate an emerging application, 3D scene understan...
We introduce SceneNet RGB-D, expanding the previous work of SceneNet to
...
This paper presents a new method for parallel-jaw grasping of isolated
o...
A multi-view image sequence provides a much richer capacity for object
r...
SLAM has matured significantly over the past few years, and is beginning...
Sculptors often deviate from geometric accuracy in order to enhance the
...
Real-time dense computer vision and SLAM offer great potential for a new...