-
Article
Open AccessA survey of uncertainty in deep neural networks
Over the last decade, neural networks have reached almost every field of science and become a crucial part of various real world applications. Due to the increasing spread, confidence in neural network predict...
-
Chapter and Conference Paper
Out-of-Distribution Detection for Adaptive Computer Vision
It is well known that computer vision can be unreliable when faced with previously unseen imaging conditions. This paper proposes a method to adapt camera parameters according to a normalizing flow-based out-o...
-
Article
Open AccessSRT3D: A Sparse Region-Based 3D Object Tracking Approach for the Real World
Region-based methods have become increasingly popular for model-based, monocular 3D tracking of texture-less objects in cluttered scenes. However, while they achieve state-of-the-art results, most methods are ...
-
Chapter and Conference Paper
Introspective Robot Perception Using Smoothed Predictions from Bayesian Neural Networks
This work focuses on improving uncertainty estimation in the field of object classification from RGB images and demonstrates its benefits in two robotic applications. We employ a Bayesian Neural Network (BNN),...
-
Chapter and Conference Paper
Robust Vision-Based Pose Correction for a Robotic Manipulator Using Active Markers
Robots with elastic or lightweight components are becoming common in research, but can suffer from undesired positioning imprecision, which motivates a vision-based pose correction of the manipulator. For robo...
-
Chapter and Conference Paper
Effective Version Space Reduction for Convolutional Neural Networks
In active learning, sampling bias could pose a serious inconsistency problem and hinder the algorithm from finding the optimal hypothesis. However, many methods for neural networks are hypothesis space agnosti...
-
Chapter and Conference Paper
A Sparse Gaussian Approach to Region-Based 6DoF Object Tracking
We propose a novel, highly efficient sparse approach to region-based 6DoF object tracking that requires only a monocular RGB camera and the 3D object model. The key contribution of our work is a probabilistic ...
-
Article
Augmented Autoencoders: Implicit 3D Orientation Learning for 6D Object Detection
We propose a real-time RGB-based pipeline for object detection and 6D pose estimation. Our novel 3D orientation estimation is based on a variant of the Denoising Autoencoder that is trained on simulated views ...
-
Chapter and Conference Paper
6DoF Pose Estimation for Industrial Manipulation Based on Synthetic Data
We present a perception system for mobile manipulation tasks. The primary design goal of the proposed system is to minimize human interaction during system setup which is achieved by several means, such as au...
-
Chapter and Conference Paper
Simultaneous Calibration and Map**
We present evaluation experiments of a hand-eye calibration and camera-camera calibration method, which is applicable to cases where classical calibration methods fail. As described in our earlier works, the c...
-
Chapter and Conference Paper
3D Scene Reconstruction from a Single Viewport
We present a novel approach to infer volumetric reconstructions from a single viewport, based only on an RGB image and a reconstructed normal image. To overcome the problem of reconstructing regions in 3D that...
-
Chapter and Conference Paper
Implicit 3D Orientation Learning for 6D Object Detection from RGB Images
We propose a real-time RGB-based pipeline for object detection and 6D pose estimation. Our novel 3D orientation estimation is based on a variant of the Denoising Autoencoder that is trained on simulated views ...
-
Chapter
Unsupervised 3D Object Discovery and Categorization for Mobile Robots
We present a method for mobile robots to learn the concept of objects and categorize them without supervision using 3D point clouds from a laser scanner as input. In particular, we address the challenges of ca...
-
Chapter
Driven Learning for Driving: How Introspection Improves Semantic Map**
This paper explores the suitability of commonly employed classification methods to action-selection tasks in robotics, and argues that a classifier’s introspective capacity is a vital but as yet largely under-app...
-
Chapter and Conference Paper
Non-rigid 3D Shape Retrieval via Large Margin Nearest Neighbor Embedding
In this paper, we propose a highly efficient metric learning approach to non-rigid 3D shape analysis. From a training set of 3D shapes from different classes, we learn a transformation of the shapes which opti...
-
Chapter
SPENCER: A Socially Aware Service Robot for Passenger Guidance and Help in Busy Airports
We present an ample description of a socially compliant mobile robotic platform, which is developed in the EU-funded project SPENCER. The purpose of this robot is to assist, inform and guide passengers in larg...
-
Chapter
A Bayesian Approach to Learning 3D Representations of Dynamic Environments
We propose a novel probabilistic approach to learning spatial representations of dynamic environments from 3D laser range measurements. Whilst most of the previous techniques developed in robotics address this...
-
Chapter and Conference Paper
Environment-Adaptive Learning: How Clustering Helps to Obtain Good Training Data
In this paper, we propose a method to combine unsupervised and semi-supervised learning (SSL) into a system that is able to adaptively learn objects in a given environment with very little user interaction. Th...
-
Chapter and Conference Paper
Active Online Learning for Interactive Segmentation Using Sparse Gaussian Processes
We present an active learning framework for image segmentation with user interaction. Our system uses a sparse Gaussian Process classifier (GPC) trained on manually labeled image pixels (user scribbles) and re...
-
Chapter and Conference Paper
Multiclass Multimodal Detection and Tracking in Urban Environments ⋆
This paper presents a novel approach to detect and track pedestrians and cars based on the combined information retrieved from a camera and a laser range scanner. Laser data points are classified using boosted...
