2,096 Hits in 3.3 sec

Point Transformer [article]

Nico Engel, Vasileios Belagiannis, Klaus Dietmayer
2020 arXiv   pre-print
In this work, we present Point Transformer, a deep neural network that operates directly on unordered and unstructured point sets. We design Point Transformer to extract local and global features and relate both representations by introducing the local-global attention mechanism, which aims to capture spatial point relations and shape information. For that purpose, we propose SortNet, as part of the Point Transformer, which induces input permutation invariance by selecting points based on a
more » ... ned score. The output of Point Transformer is a sorted and permutation invariant feature list that can directly be incorporated into common computer vision applications. We evaluate our approach on standard classification and part segmentation benchmarks to demonstrate competitive results compared to the prior work.
arXiv:2011.00931v1 fatcat:4wc4zcgiibdwdiqa5ac4v44rje

Attention-based Vehicle Self-Localization with HD Feature Maps [article]

Nico Engel, Vasileios Belagiannis, Klaus Dietmayer
2021 arXiv   pre-print
We present a vehicle self-localization method using point-based deep neural networks. Our approach processes measurements and point features, i.e. landmarks, from a high-definition digital map to infer the vehicle's pose. To learn the best association and incorporate local information between the point sets, we propose an attention mechanism that matches the measurements to the corresponding landmarks. Finally, we use this representation for the point-cloud registration and the subsequent pose
more » ... egression task. Furthermore, we introduce a training simulation framework that artificially generates measurements and landmarks to facilitate the deployment process and reduce the cost of creating extensive datasets from real-world data. We evaluate our method on our dataset, as well as an adapted version of the Kitti odometry dataset, where we achieve superior performance compared to related approaches; and additionally show dominant generalization capabilities.
arXiv:2107.07787v1 fatcat:ky3dx5b5ujesplfghuesn4pl7e

Deep Object Tracking on Dynamic Occupancy Grid Maps Using RNNs [article]

Nico Engel, Stefan Hoermann, Philipp Henzler, Klaus Dietmayer
2018 arXiv   pre-print
The comprehensive representation and understanding of the driving environment is crucial to improve the safety and reliability of autonomous vehicles. In this paper, we present a new approach to establish an environment model containing a segmentation between static and dynamic background and parametric modeled objects with shape, position and orientation. Multiple laser scanners are fused into a dynamic occupancy grid map resulting in a 360 perception of the environment. A single-stage deep
more » ... volutional neural network is combined with a recurrent neural network, which takes a time series of the occupancy grid map as input and tracks cell states and its corresponding object hypotheses. The labels for training are created unsupervised with an automatic label generation algorithm. The proposed methods are evaluated in real-world experiments in complex inner city scenarios using the aforementioned 360 laser perception. The results show a better object detection accuracy in comparison with our old approach as well as an AUC score of 0.946 for the dynamic and static segmentation. Furthermore, we gain an improved detection for occluded objects and a more consistent size estimation due to the usage of time series as input and the memory about previous states introduced by the recurrent neural network.
arXiv:1805.08986v1 fatcat:ejettusx45e47ezptgvhmycmli

The risk appetite of private equity sponsors

Reiner Braun, Nico Engel, Peter Hieber, Rudi Zagst
2011 Journal of Empirical Finance  
Using a unique proprietary data set of 460 realized buyouts completed between 1990 and 2005, we examine the risk appetite of private equity (PE) sponsors in different states of the PE market and analyse key determinants of deal-level equity risk. We develop a new approach to mathematically model PE investment equity risk based on the Black-Cox default model. We find higher equity volatilities during boom periods. Further, deals conducted by more reputed PE sponsors have lower equity
more » ... as they are unwilling to imperil their reputation by taking excessive risks. In addition, we find that PE sponsors' risk appetite is negatively related to the ownership stake in the buyout target company.
doi:10.1016/j.jempfin.2011.07.002 fatcat:ipjrrihksvbmvgx7rvdvyay3gm

DeepLocalization: Landmark-based Self-Localization with Deep Neural Networks [article]

Nico Engel, Stefan Hoermann, Markus Horn, Vasileios Belagiannis, Klaus Dietmayer
2019 arXiv   pre-print
We address the problem of vehicle self-localization from multi-modal sensor information and a reference map. The map is generated off-line by extracting landmarks from the vehicle's field of view, while the measurements are collected similarly on the fly. Our goal is to determine the autonomous vehicle's pose from the landmark measurements and map landmarks. To learn this mapping, we propose DeepLocalization, a deep neural network that regresses the vehicle's translation and rotation parameters
more » ... from unordered and dynamic input landmarks. The proposed network architecture is robust to changes of the dynamic environment and can cope with a small number of extracted landmarks. During the training process we rely on synthetically generated ground-truth. In our experiments, we evaluate two inference approaches in real-world scenarios. We show that DeepLocalization can be combined with regular GPS signals and filtering algorithms such as the extended Kalman filter. Our approach achieves state-of-the-art accuracy and is about ten times faster than the related work.
arXiv:1904.09007v2 fatcat:wkhoqqawufashjvwjoc655mpki

Bioinspired synthesis of mesoporous silicas

Qianyao Sun, Engel G Vrieling, Rutger A van Santen, Nico A.J.M Sommerdijk
2004 Current opinion in solid state & materials science  
Recent years have witnessed rapid growth in the number of new investigations at the interface of materials chemistry and biology. This review highlights the recent developments in the studies of protein-mediated silica biomineralization in diatoms and the "downscaling" and "upscaling" models derived thereof, as well as the recent progress in the fabrication of artificial silicas with novel pore structures and morphologies at different length scales based on these new insights into biosilica formation.
doi:10.1016/j.cossms.2004.01.005 fatcat:aul2mwqylbh5xeq2nnbg6ygdrm

Domain structure of secretin PulD revealed by limited proteolysis and electron microscopy

Nico Nouwen, Henning Stahlberg, Anthony P. Pugsley, Andreas Engel
2000 EMBO Journal  
doi:10.1093/emboj/19.10.2229 pmid:10811614 pmcid:PMC384361 fatcat:kp6mqw5sm5hpzas3jamv6dqugi

Can higher-level phylogenies of weevils explain their evolutionary success? A critical review

2010 Systematic Entomology  
Correspondence: Nico M. Franz, Department of Biology, University of Puerto Rico, Call Box 9000, Mayagüez, PR 00681-9000, U.S.A.  ...  The timing of weevil diversification is prima facie not that different from other nonherbivorous beetle lineages (Zherikhin, 2002; Grimaldi & Engel, 2005; Hunt et al., 2007) .  ... 
doi:10.1111/j.1365-3113.2010.00534.x fatcat:xu5ngb5cuve67j6metapcmuyem

Supplement_Material – Supplemental material for Subgroups with typical courses of depressive symptoms in an elderly population during 13 years of observation: Results from the Heinz Nixdorf Recall Study

Miriam Engel, Karl-Heinz Jöckel, Nico Dragano, Miriam Engels, Susanne Moebus
2020 Figshare  
, Karl-Heinz Jöckel, Nico Dragano, Miriam Engels and Susanne Moebus in International Journal of Social Psychiatry  ...  material, Supplement_Material for Subgroups with typical courses of depressive symptoms in an elderly population during 13 years of observation: Results from the Heinz Nixdorf Recall Study by Miriam Engel  ... 
doi:10.25384/sage.12571505 fatcat:sfeaanvsuvctzdwl4kznkise4q

The Gender Lens: Development of a learning aid to introduce gender medicine

Simone Weyers, Anja Vervoorts, Nico Dragano, Miriam Engels
2017 GMS Journal for Medical Education  
Dragano 3 Studierende der Medizin sollten in der Lehre frühzeitig mit solchen Miriam Engels 3 Unterschieden vertraut gemacht werden.  ...  Gendermedizin Zusammenfassung Hintergrund und Ziel: Gendermedizin berücksichtigt biologische und soziale Unterschiede zwischen Männern und Frauen bei der Häufigkeit Simone Weyers 1 Anja Vervoorts 2 Nico  ... 
doi:10.3205/zma001094 pmid:28584865 pmcid:PMC5450432 fatcat:fatwb22cwvgthhokuhhu32chsq

Chemosensory Event-Related Potentials in Response to Nasal Propylene Glycol Stimulation

Mohammad Sirous, Nico Sinning, Till R. Schneider, Uwe Friese, Jürgen Lorenz, Andreas K. Engel
2019 Frontiers in Human Neuroscience  
Propylene glycol, also denoted as 1.2 propanediol (C3H8O2), often serves as a solvent for dilution of olfactory stimuli. It is supposed to serve as a neutral substance and has been used in many behavioral and electrophysiological studies to dilute pure olfactory stimuli. However, the effect of propylene glycol on perception and on neuronal responses has hitherto never been studied. In this study we tested by means of a threshold test, whether a nasal propylene glycol stimulation is recognizable
more » ... by humans. Participants were able to recognize propylene glycol at a threshold of 42% concentration and reported a slight cooling effect. In addition to the threshold test, we recorded electroencephalography (EEG) during nasal propylene glycol stimulation to study the neuronal processing of the stimulus. We used a flow olfactometer and stimulated 15 volunteers with three different concentrations of propylene glycol (40 trials each) and water as a control condition (40 trials). To evaluate the neuronal response, we analyzed the event-related potentials (ERPs) and power modulations. The task of the volunteers was to identify a change (olfactory, thermal, or tactile) in the continuous air flow generated by the flow olfactometer. The analysis of the ERPs showed that propylene glycol generates a clear P2 component, which was also visible in the frequency domain as an evoked power response in the theta-band. The source analysis of the P2 revealed a widespread involvement of brain regions, including the postcentral gyrus, the insula and adjacent operculum, the thalamus, and the cerebellum. Thus, it is possible that trigeminal stimulation can at least partly account for sensations and brain responses elicited by propylene glycol. Based on these results, we conclude that the use of high propylene glycol concentrations to dilute fragrances complicates the interpretation of presumed purely olfactory effects.
doi:10.3389/fnhum.2019.00099 pmid:30949040 pmcid:PMC6435593 fatcat:mjrbm3olkzblpjzzbe4gm6ot2q

DeepCLR: Correspondence-Less Architecture for Deep End-to-End Point Cloud Registration [article]

Markus Horn, Nico Engel, Vasileios Belagiannis, Michael Buchholz, Klaus Dietmayer
2020 arXiv   pre-print
This work addresses the problem of point cloud registration using deep neural networks. We propose an approach to predict the alignment between two point clouds with overlapping data content, but displaced origins. Such point clouds originate, for example, from consecutive measurements of a LiDAR mounted on a moving platform. The main difficulty in deep registration of raw point clouds is the fusion of template and source point cloud. Our proposed architecture applies flow embedding to tackle
more » ... is problem, which generates features that describe the motion of each template point. These features are then used to predict the alignment in an end-to-end fashion without extracting explicit point correspondences between both input clouds. We rely on the KITTI odometry and ModelNet40 datasets for evaluating our method on various point distributions. Our approach achieves state-of-the-art accuracy and the lowest run-time of the compared methods.
arXiv:2007.11255v1 fatcat:vvilbaqk5vfzbaoifsba3of3sy

Triple Graph Grammars in the Large for Translating Satellite Procedures [chapter]

Frank Hermann, Susann Gottmann, Nico Nachtigall, Hartmut Ehrig, Benjamin Braatz, Gianluigi Morelli, Alain Pierre, Thomas Engel, Claudia Ermel
2014 Lecture Notes in Computer Science  
Software translation is a challenging task. Several requirements are important -including automation of the execution, maintainability of the translation patterns, and, most importantly, reliability concerning the correctness of the translation. Triple graph grammars (TGGs) have shown to be an intuitive, welldefined technique for model translation. In this paper, we leverage TGGs for industry scale software translations. The approach is implemented using the Eclipse-based graph transformation
more » ... ol Henshin and has been successfully applied in a large industrial project with the satellite operator SES on the translation of satellite control procedures. We evaluate the approach regarding requirements from the project and performance on a complete set of procedures of one satellite. 4 Awk Community: 5 In [15], we present a short overview of the PIL2SPELL project.
doi:10.1007/978-3-319-08789-4_9 fatcat:lygocqupfbge3jfwscxadu4nwm

Local-scale changes in plant diversity: reassessments and implications for biodiversity–ecosystem function experiments

Alexandra Wright, Markus Bernhardt-Römermann, Dylan Craven, Anne Ebeling, Jan Engel, Jes Hines, Nico Eisenhauer
2014 Proceedings of Peerage of Science  
doi:10.14726/procpos.2014.e6 fatcat:cwelpgnr3falrfzsclps3cvqki

Using Sensorimotor Contingencies for Terrain Discrimination and Adaptive Walking Behavior in the Quadruped Robot Puppy [chapter]

Matej Hoffmann, Nico M. Schmidt, Rolf Pfeifer, Andreas K. Engel, Alexander Maye
2012 Lecture Notes in Computer Science  
In conventional "sense-think-act" control architectures, perception is reduced to a passive collection of sensory information, followed by a mapping onto a prestructured internal world model. For biological agents, Sensorimotor Contingency Theory (SMCT) posits that perception is not an isolated processing step, but is constituted by knowing and exercising the law-like relations between actions and resulting changes in sensory stimulation. We present a computational model of SMCT for controlling
more » ... the behavior of a quadruped robot running on different terrains. Our experimental study demonstrates that: (i) Sensory-Motor Contingencies (SMC) provide better discrimination capabilities of environmental properties than conventional recognition from the sensory signals alone; (ii) discrimination is further improved by considering the action context on a longer time scale; (iii) the robot can utilize this knowledge to adapt its behavior for maximizing its stability.
doi:10.1007/978-3-642-33093-3_6 fatcat:tr5aikpr4jcdrc7yakng6r57yi
« Previous Showing results 1 — 15 out of 2,096 results