Search Results for author: Rudolph Triebel

Found 40 papers, 17 papers with code

3D Scene Reconstruction from a Single Viewport

1 code implementation ECCV 2020 Maximilian Denninger, Rudolph Triebel

To overcome the problem of reconstructing regions in 3D that are occluded in the 2D image, we propose to learn this information from synthetically generated high-resolution data.

3D Scene Reconstruction

Estimating Model Uncertainty of Neural Network in Sparse Information Form

no code implementations ICML 2020 Jongseok Lee, Matthias Humt, Jianxiang Feng, Rudolph Triebel

As a result, we show that the information form of MND can be scalably applied to represent model uncertainty in MND.

Dimensionality Reduction

Topology-Matching Normalizing Flows for Out-of-Distribution Detection in Robot Learning

no code implementations11 Nov 2023 Jianxiang Feng, JongSeok Lee, Simon Geisler, Stephan Gunnemann, Rudolph Triebel

To facilitate reliable deployments of autonomous robots in the real world, Out-of-Distribution (OOD) detection capabilities are often required.

Density Estimation object-detection +3

Learning Expressive Priors for Generalization and Uncertainty Estimation in Neural Networks

1 code implementation15 Jul 2023 Dominik Schnaus, JongSeok Lee, Daniel Cremers, Rudolph Triebel

In this work, we propose a novel prior learning method for advancing generalization and uncertainty estimation in deep neural networks.

Continual Learning Generalization Bounds

Density-based Feasibility Learning with Normalizing Flows for Introspective Robotic Assembly

1 code implementation3 Jul 2023 Jianxiang Feng, Matan Atad, Ismael Rodríguez, Maximilian Durner, Stephan Günnemann, Rudolph Triebel

Machine Learning (ML) models in Robotic Assembly Sequence Planning (RASP) need to be introspective on the predicted solutions, i. e. whether they are feasible or not, to circumvent potential efficiency degradation.

Out of Distribution (OOD) Detection

Out-of-Distribution Detection for Adaptive Computer Vision

no code implementations16 May 2023 Simon Kristoffersson Lind, Rudolph Triebel, Luigi Nardi, Volker Krueger

It is well known that computer vision can be unreliable when faced with previously unseen imaging conditions.

Out-of-Distribution Detection

6D Object Pose Estimation from Approximate 3D Models for Orbital Robotics

no code implementations23 Mar 2023 Maximilian Ulmer, Maximilian Durner, Martin Sundermeyer, Manuel Stoiber, Rudolph Triebel

We present a novel technique to estimate the 6D pose of objects from single images where the 3D geometry of the object is only given approximately and not as a precise 3D model.

6D Pose Estimation using RGB

Efficient and Feasible Robotic Assembly Sequence Planning via Graph Representation Learning

1 code implementation17 Mar 2023 Matan Atad, Jianxiang Feng, Ismael Rodríguez, Maximilian Durner, Rudolph Triebel

With GRACE, we are able to extract meaningful information from the graph input and predict assembly sequences in a step-by-step manner.

Graph Representation Learning

RECALL: Rehearsal-free Continual Learning for Object Classification

1 code implementation29 Sep 2022 Markus Knauer, Maximilian Denninger, Rudolph Triebel

Our approach is called RECALL, as the network recalls categories by calculating logits for old categories before training new ones.

Classification Continual Learning +1

A Multi-body Tracking Framework -- From Rigid Objects to Kinematic Structures

1 code implementation2 Aug 2022 Manuel Stoiber, Martin Sundermeyer, Wout Boerdijk, Rudolph Triebel

Our approach focuses on methods that employ Newton-like optimization techniques, which are widely used in object tracking.

3D Object Tracking 6D Pose Estimation +2

A Model for Multi-View Residual Covariances based on Perspective Deformation

no code implementations1 Feb 2022 Alejandro Fontan, Laura Oliva, Javier Civera, Rudolph Triebel

In this work, we derive a model for the covariance of the visual residuals in multi-view SfM, odometry and SLAM setups.

Making Curiosity Explicit in Vision-based RL

no code implementations28 Sep 2021 Elie Aljalbout, Maximilian Ulmer, Rudolph Triebel

Our method enhances the exploration capability of the RL algorithms by taking advantage of the SRL setup.

Reinforcement Learning (RL) Representation Learning

Introspective Robot Perception using Smoothed Predictions from Bayesian Neural Networks

no code implementations27 Sep 2021 Jianxiang Feng, Maximilian Durner, Zoltan-Csaba Marton, Ferenc Balint-Benczedi, Rudolph Triebel

This work focuses on improving uncertainty estimation in the field of object classification from RGB images and demonstrates its benefits in two robotic applications.

Domain Adaptation

Bayesian Active Learning for Sim-to-Real Robotic Perception

no code implementations23 Sep 2021 Jianxiang Feng, JongSeok Lee, Maximilian Durner, Rudolph Triebel

While learning from synthetic training data has recently gained an increased attention, in real-world robotic applications, there are still performance deficiencies due to the so-called Sim-to-Real gap.

Active Learning Informativeness +1

Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes

no code implementations20 Sep 2021 JongSeok Lee, Jianxiang Feng, Matthias Humt, Marcus G. Müller, Rudolph Triebel

This paper presents a probabilistic framework to obtain both reliable and fast uncertainty estimates for predictions with Deep Neural Networks (DNNs).

Gaussian Processes object-detection +1

Towards Robust Monocular Visual Odometry for Flying Robots on Planetary Missions

1 code implementation12 Sep 2021 Martin Wudenka, Marcus G. Müller, Nikolaus Demmel, Armin Wedler, Rudolph Triebel, Daniel Cremers, Wolfgang Stürzl

In contrast to most other approaches, our framework can also handle rotation-only motions that are particularly challenging for monocular odometry systems.

Monocular Visual Odometry Optical Flow Estimation

A Survey of Uncertainty in Deep Neural Networks

no code implementations7 Jul 2021 Jakob Gawlikowski, Cedrique Rovile Njieutcheu Tassi, Mohsin Ali, JongSeok Lee, Matthias Humt, Jianxiang Feng, Anna Kruspe, Rudolph Triebel, Peter Jung, Ribana Roscher, Muhammad Shahzad, Wen Yang, Richard Bamler, Xiao Xiang Zhu

Different examples from the wide spectrum of challenges in different fields give an idea of the needs and challenges regarding uncertainties in practical applications.

Data Augmentation

Contact-GraspNet: Efficient 6-DoF Grasp Generation in Cluttered Scenes

1 code implementation25 Mar 2021 Martin Sundermeyer, Arsalan Mousavian, Rudolph Triebel, Dieter Fox

Our novel grasp representation treats 3D points of the recorded point cloud as potential grasp contacts.

Grasp Generation Robotic Grasping

Unknown Object Segmentation from Stereo Images

2 code implementations11 Mar 2021 Maximilian Durner, Wout Boerdijk, Martin Sundermeyer, Werner Friedl, Zoltan-Csaba Marton, Rudolph Triebel

This has the major advantage that instead of a noisy, and potentially incomplete depth map as an input, on which the segmentation is computed, we use the original image pair to infer the object instances and a dense depth map.

Instance Segmentation Segmentation +1

Learning to Localize in New Environments from Synthetic Training Data

1 code implementation9 Nov 2020 Dominik Winkelbauer, Maximilian Denninger, Rudolph Triebel

Our approach outperforms the 5-point algorithm using SIFT features on equally big images and additionally surpasses all previous learning-based approaches that were trained on different data.

Visual Localization

"What's This?" -- Learning to Segment Unknown Objects from Manipulation Sequences

1 code implementation6 Nov 2020 Wout Boerdijk, Martin Sundermeyer, Maximilian Durner, Rudolph Triebel

Furthermore, while the motion of the manipulator and the object are substantial cues for our algorithm, we present means to robustly deal with distraction objects moving in the background, as well as with completely static scenes.

Foreground Segmentation Segmentation +1

Bayesian Optimization Meets Laplace Approximation for Robotic Introspection

no code implementations30 Oct 2020 Matthias Humt, JongSeok Lee, Rudolph Triebel

In robotics, deep learning (DL) methods are used more and more widely, but their general inability to provide reliable confidence estimates will ultimately lead to fragile and unreliable systems.

Bayesian Optimization

DOT: Dynamic Object Tracking for Visual SLAM

no code implementations30 Sep 2020 Irene Ballester, Alejandro Fontan, Javier Civera, Klaus H. Strobl, Rudolph Triebel

In this paper we present DOT (Dynamic Object Tracking), a front-end that added to existing SLAM systems can significantly improve their robustness and accuracy in highly dynamic environments.

Instance Segmentation Object Tracking +2

Gaussian Process Gradient Maps for Loop-Closure Detection in Unstructured Planetary Environments

no code implementations1 Sep 2020 Cedric Le Gentil, Mallikarjuna Vayugundla, Riccardo Giubilato, Wolfgang Stürzl, Teresa Vidal-Calleja, Rudolph Triebel

Loop closures are verified by leveraging both the spatial characteristic of the elevation maps (SE(2) registration) and the probabilistic nature of the GP representation.

Image Registration Loop Closure Detection +1

Learning Multiplicative Interactions with Bayesian Neural Networks for Visual-Inertial Odometry

no code implementations15 Jul 2020 Kashmira Shinde, Jong-Seok Lee, Matthias Humt, Aydin Sezgin, Rudolph Triebel

This paper presents an end-to-end multi-modal learning approach for monocular Visual-Inertial Odometry (VIO), which is specifically designed to exploit sensor complementarity in the light of sensor degradation scenarios.

Inductive Bias

Effective Version Space Reduction for Convolutional Neural Networks

no code implementations22 Jun 2020 Jiayu Liu, Ioannis Chiotellis, Rudolph Triebel, Daniel Cremers

In active learning, sampling bias could pose a serious inconsistency problem and hinder the algorithm from finding the optimal hypothesis.

Active Learning Image Classification

Estimating Model Uncertainty of Neural Networks in Sparse Information Form

no code implementations20 Jun 2020 Jongseok Lee, Matthias Humt, Jianxiang Feng, Rudolph Triebel

We present a sparse representation of model uncertainty for Deep Neural Networks (DNNs) where the parameter posterior is approximated with an inverse formulation of the Multivariate Normal Distribution (MND), also known as the information form.

Dimensionality Reduction

Segmentation of Surgical Instruments for Minimally-Invasive Robot-Assisted Procedures Using Generative Deep Neural Networks

no code implementations5 Jun 2020 Iñigo Azqueta-Gavaldon, Florian Fröhlich, Klaus Strobl, Rudolph Triebel

Nevertheless, one of the caveats of this approach is that the model is unable to generalize well to other surgical instruments with a different shape from the one used for training.

Domain Adaptation Semantic Segmentation

Representing Model Uncertainty of Neural Networks in Sparse Information Form

no code implementations25 Sep 2019 JongSeok Lee, Rudolph Triebel

This paper addresses the problem of representing a system's belief using multi-variate normal distributions (MND) where the underlying model is based on a deep neural network (DNN).

Non-Parametric Calibration for Classification

1 code implementation12 Jun 2019 Jonathan Wenger, Hedvig Kjellström, Rudolph Triebel

Many applications of classification methods not only require high accuracy but also reliable estimation of predictive uncertainty.

Active Learning Classification +2

Learning to Drive using Inverse Reinforcement Learning and Deep Q-Networks

no code implementations12 Dec 2016 Sahand Sharifzadeh, Ioannis Chiotellis, Rudolph Triebel, Daniel Cremers

We propose an inverse reinforcement learning (IRL) approach using Deep Q-Networks to extract the rewards in problems with large state spaces.

Autonomous Driving reinforcement-learning +1

Cannot find the paper you are looking for? You can Submit a new open access paper.