no code implementations • 26 Feb 2024 • Benjamin Alt, Florian Stöckl, Silvan Müller, Christopher Braun, Julian Raible, Saad Alhasan, Oliver Rettig, Lukas Ringle, Darko Katic, Rainer Jäkel, Michael Beetz, Marcus Strand, Marco F. Huber
Surface treatment tasks such as grinding, sanding or polishing are a vital step of the value chain in many industries, but are notoriously challenging to automate.
1 code implementation • 13 Dec 2023 • Marc-André Zöller, Marius Lindauer, Marco F. Huber
The framework employs Bayesian optimization, to automatically construct pipelines from statistical, machine learning (ML) and deep neural network (DNN) models.
no code implementations • 7 Nov 2023 • Ruyu Wang, Sabrina Schmedding, Marco F. Huber
Although looking appealing to human eyes, training a model on purely synthetic images for downstream image processing tasks like image classification often results in an undesired performance drop compared to training on real data.
no code implementations • 21 Jul 2023 • Danilo Brajovic, Niclas Renner, Vincent Philipp Goebels, Philipp Wagner, Benjamin Fresz, Martin Biller, Mara Klaeb, Janika Kutz, Jens Neuhuettler, Marco F. Huber
In this work we merge recent regulation efforts by the European Union and first proposals for AI guidelines with recent trends in research: data and model cards.
no code implementations • 6 Jul 2023 • Christian Jauch, Timo Leitritz, Marco F. Huber
The pipeline consists of a general machine learning model for hand pose estimation trained on a generalized dataset, spatial and temporal filtering to account for anatomical constraints of the hand, and a retraining step to improve the model.
no code implementations • 21 Jun 2023 • Marc-André Zöller, Fabian Mauthe, Peter Zeiler, Marius Lindauer, Marco F. Huber
Recently, data-driven approaches to RUL predictions are becoming prevalent over model-based approaches since no underlying physical knowledge of the engineering system is required.
no code implementations • 21 May 2023 • Xinyang Wu, Elisabeth Wedernikow, Christof Nitsche, Marco F. Huber
In recent years, the development of Artificial Intelligence (AI) has shown tremendous potential in diverse areas.
no code implementations • 16 Feb 2023 • Ruyu Wang, Sabrina Hoppe, Eduardo Monari, Marco F. Huber
Data-hunger and data-imbalance are two major pitfalls in many deep learning approaches.
1 code implementation • 26 Nov 2022 • Simeon Brüggenjürgen, Nina Schaaf, Pascal Kerschke, Marco F. Huber
This work introduces a novel interpretable machine learning method called Mixture of Decision Trees (MoDT).
no code implementations • 7 Oct 2022 • Tobias Nagel, Marco F. Huber
Identifying parameters in a system of nonlinear, ordinary differential equations is vital for designing a robust controller.
no code implementations • 23 Aug 2022 • Paul-Amaury Matt, Rosina Ziegler, Danilo Brajovic, Marco Roth, Marco F. Huber
Our goal in this paper is to automatically extract a set of decision rules (rule set) that best explains a classification data set.
no code implementations • 9 Mar 2022 • Tassneem Helal, Fady Aziz, Omar Metwally, Marco F. Huber, Dominik Alscher, Christoph Wasser, Urs Schneider
Estimating human vital signs in a contactless non-invasive method using radar provides a convenient method in the medical field to conduct several health checkups easily and quickly.
1 code implementation • 24 Feb 2022 • Marc-André Zöller, Waldemar Titov, Thomas Schlegel, Marco F. Huber
Even though such automatically synthesized ML pipelines are able to achieve a competitive performance, recent studies have shown that users do not trust models constructed by AutoML due to missing transparency of AutoML systems and missing explanations for the constructed ML pipelines.
1 code implementation • 21 Feb 2022 • Raoul Schönhof, Jannes Elstner, Radu Manea, Steffen Tauber, Ramez Awad, Marco F. Huber
In this work, we present a deep residual 3D autoencoder based on the EfficientNet architecture, intended for transfer learning tasks related to 3D CAD model assessment.
no code implementations • 2 Nov 2021 • Fady Aziz, Omar Metwally, Pascal Weller, Urs Schneider, Marco F. Huber
In this paper, a MIMO radar is used to formulate a novel micro-motion spectrogram for the angular velocity ({\mu}-{\omega}) in non-tangential scenarios.
no code implementations • 16 Oct 2021 • Pascal Weller, Fady Aziz, Sherif Abdulatif, Urs Schneider, Marco F. Huber
Radar for deep learning-based human identification has become a research area of increasing interest.
no code implementations • 3 Oct 2021 • Philipp Wagner, Xinyang Wu, Marco F. Huber
Compared to point estimates calculated by standard neural networks, Bayesian neural networks (BNN) provide probability distributions over the output predictions and model parameters, i. e., the weights.
no code implementations • 3 Oct 2021 • Kilian Kleeberger, Jonathan Schnitzler, Muhammad Usman Khalid, Richard Bormann, Werner Kraus, Marco F. Huber
This paper introduces a novel approach for the grasping and precise placement of various known rigid objects using multiple grippers within highly cluttered scenes.
1 code implementation • 27 Aug 2021 • Raphael Lamprecht, Ferdinand Wurst, Marco F. Huber
Maintenance scheduling is a complex decision-making problem in the production domain, where a number of maintenance tasks and resources has to be assigned and scheduled to production entities in order to prevent unplanned production downtime.
no code implementations • 1 Jul 2021 • Nina Schaaf, Omar de Mitri, Hang Beom Kim, Alexander Windberger, Marco F. Huber
For this purpose, first an artificial dataset with a known bias is created and used to train intentionally biased CNNs.
no code implementations • 30 May 2021 • Fady Aziz, Bassam Elmakhzangy, Christophe Maufroy, Urs Schneider, Marco F. Huber
The algorithm is implemented on a microcontroller subsystem of the radar kit to qualify the perception system for embedded integration in powered prosthetic legs.
no code implementations • 15 Apr 2021 • Kilian Kleeberger, Markus Völk, Richard Bormann, Marco F. Huber
Single shot approaches have demonstrated tremendous success on various computer vision tasks.
no code implementations • 26 Jan 2021 • Marc-André Zöller, Tien-Dung Nguyen, Marco F. Huber
We prove the effectiveness and competitiveness of our approach on 28 data sets used in well-established AutoML benchmarks in comparison with state-of-the-art AutoML frameworks.
no code implementations • 12 Jan 2021 • Kilian Kleeberger, Markus Völk, Marius Moosmann, Erik Thiessenhusen, Florian Roth, Richard Bormann, Marco F. Huber
In this paper, we introduce a novel learning-based approach for grasping known rigid objects in highly cluttered scenes and precisely placing them based on depth images.
no code implementations • 16 Nov 2020 • Nadia Burkart, Marco F. Huber
This survey paper provides essential definitions, an overview of the different principles and methodologies of explainable Supervised Machine Learning (SML).
no code implementations • 3 Sep 2020 • Marco F. Huber
In this paper a novel approach towards fully Bayesian NNs is proposed, where training and predictions of a perceptron are performed within the Bayesian inference framework in closed-form.
no code implementations • 27 Apr 2020 • Kilian Kleeberger, Marco F. Huber
In this paper, we introduce a novel single shot approach for 6D object pose estimation of rigid objects based on depth images.
1 code implementation • 6 Dec 2019 • Kilian Kleeberger, Christian Landgraf, Marco F. Huber
In this paper, we introduce a new public dataset for 6D object pose estimation and instance segmentation for industrial bin-picking.
no code implementations • 8 Sep 2019 • Muhammad Usman Khalid, Janik M. Hager, Werner Kraus, Marco F. Huber, Marc Toussaint
For most industrial bin picking solutions, the pose of a workpiece is localized by matching a CAD model to point cloud obtained from 3D sensor.
1 code implementation • 26 Apr 2019 • Marc-André Zöller, Marco F. Huber
This paper is a combination of a survey on current AutoML methods and a benchmark of popular AutoML frameworks on real data sets.
no code implementations • 10 Apr 2019 • Nina Schaaf, Marco F. Huber, Johannes Maucher
One obstacle that so far prevents the introduction of machine learning models primarily in critical areas is the lack of explainability.
no code implementations • 20 Mar 2012 • Marc Peter Deisenroth, Ryan Turner, Marco F. Huber, Uwe D. Hanebeck, Carl Edward Rasmussen
We propose a principled algorithm for robust Bayesian filtering and smoothing in nonlinear stochastic dynamic systems when both the transition function and the measurement function are described by non-parametric Gaussian process (GP) models.