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.
As a result, we show that the information form of MND can be scalably applied to represent model uncertainty in MND.
To facilitate reliable deployments of autonomous robots in the real world, Out-of-Distribution (OOD) detection capabilities are often required.
In this work, we propose a novel prior learning method for advancing generalization and uncertainty estimation in deep neural networks.
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.
It is well known that computer vision can be unreliable when faced with previously unseen imaging conditions.
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.
With GRACE, we are able to extract meaningful information from the graph input and predict assembly sequences in a step-by-step manner.
In this work, we develop a multi-modality tracker that fuses information from visual appearance and geometry to estimate object poses.
Ranked #1 on 6D Pose Estimation on YCB-Video
This article presents a novel telepresence system for advancing aerial manipulation in dynamic and unstructured environments.
Our approach is called RECALL, as the network recalls categories by calculating logits for old categories before training new ones.
Ranked #1 on Classification on HOWS long
Our approach focuses on methods that employ Newton-like optimization techniques, which are widely used in object tracking.
Ranked #1 on 3D Object Tracking on RTB
Tracking objects in 3D space and predicting their 6DoF pose is an essential task in computer vision.
Ranked #2 on 6D Pose Estimation on OPT
In this work, we derive a model for the covariance of the visual residuals in multi-view SfM, odometry and SLAM setups.
Finally, we use a pre-rendered sparse viewpoint model to create a joint posterior probability for the object pose.
Our method enhances the exploration capability of RL algorithms, by taking advantage of the SRL setup.
Our method enhances the exploration capability of the RL algorithms by taking advantage of the SRL setup.
This work focuses on improving uncertainty estimation in the field of object classification from RGB images and demonstrates its benefits in two robotic applications.
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.
This paper presents a probabilistic framework to obtain both reliable and fast uncertainty estimates for predictions with Deep Neural Networks (DNNs).
In contrast to most other approaches, our framework can also handle rotation-only motions that are particularly challenging for monocular odometry systems.
no code implementations • 7 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.
Our novel grasp representation treats 3D points of the recorded point cloud as potential grasp contacts.
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.
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.
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.
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.
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.
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.
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.
In active learning, sampling bias could pose a serious inconsistency problem and hinder the algorithm from finding the optimal hypothesis.
Ranked #107 on Image Classification on STL-10
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.
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.
This paper presents a novel telepresence system for enhancing aerial manipulation capabilities.
Accurate object segmentation is a crucial task in the context of robotic manipulation.
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).
We introduce a scalable approach for object pose estimation trained on simulated RGB views of multiple 3D models together.
Many applications of classification methods not only require high accuracy but also reliable estimation of predictive uncertainty.
Our novel 3D orientation estimation is based on a variant of the Denoising Autoencoder that is trained on simulated views of a 3D model using Domain Randomization.
Ranked #1 on 6D Pose Estimation using RGBD on T-LESS
We propose an inverse reinforcement learning (IRL) approach using Deep Q-Networks to extract the rewards in problems with large state spaces.