http://www.tu-ilmenau.de

Logo TU Ilmenau

Contact Person

Univ.-Prof. Dr.-Ing. Horst-Michael Groß

Head of department

Phone +49 3677 692858

Send email

INHALTE

Dr.-Ing. Markus Eisenbach

Contact


  Room: Zusebau 3050

  Phone: +49 (0)3677 69-4169

  Fax: +49 (0)3677 69-1665

  markus.eisenbach@tu-ilmenau.de

Vita & Research

Vita

Since 05/2019Head of the junior research group "Machine Learning" as part of the project E4SM
Since 11/2009Scientific staff member at the Department of Neuroinformatics and Cognitive Robotics
2007Internship Fraunhofer IGD Darmstadt
2004 - 2009Study of computer science at Ilmenau University of Technology

 

Research topics

  • Main research topic: Trustworthy Machine Learning
  • Machine Learning: Deep Learning, Uncertainty Estimation, Out-of-Distribution Detection, Representation Learning, Feature Selection, Distance Metric Learning, Classification
  • Computer Vision: Appearance-based Person Re-Identification, Visual Features, Illumination Compensation, Background Subtraction
  • Mobile Robotics: Object Recognition, Tracking, User Re-Identification
  • Applications: Multi-Camera Tracking, Following & Guiding People with a Mobile Robot, Automatic Inspection of Public Infrastructure, Industrical Manufacturing Processes

Research projects

  • APFel (2010-2014):
    • Video analysis tool for assisting human operators in airport surveillance
    • Main contribution: Person re-identification for multi-camera tracking
  • ROREAS (2014-2016)
    • Robot-assisted follow-up care of stroke patients
    • Main contribution: User recognition to resolve ambiguities in tracking
  • ASINVOS (2016-2018)
    • Analysis tool for distress detection in inspection data of public infrastructure
    • Main contribution: Damage spot detection by deep learning
  • ASFaLT (2018-2019)
    • Follow up project of ASINVOS
    • Goal: Fully automated tool for distress detection in inspection data of public infrastructure
    • Main contribution: Neural network calibration, uncertainty estimation
  • E4SM (2019-current)
    • Machine learning in industrical manufacturing processes
    • Head of junior research group "Machine Learning", mentoring of 2 PhD students
    • Main research contribution: Uncertainty estimation, out-of-distribution detection in deep neural networks

Teaching

Sprechzeit

Mittwochs 15:00 - 15:30 Uhr

Vorlesungen

Deep Learning for Computer Vision

(Wintersemester 2019/20)

zusammen mit D. Seichter, H.-M. Groß

Zeit: Fr. (G) 13:00 Uhr

Inhalte

  • Was ist Deep Learning? Was kann Deep Learning? Was noch nicht?
  • Frameworks
  • Grundlagen zu Neuronalen Netzwerken
    • Aufbau Neuronales Netzwerk
    • Ausgabefunktionen
    • Gewichtsinitialisierung
    • Error-Backpropagation
    • Optimierungsverfahren
    • Regularisierung
    • Multi-Layer Perceptron
    • Convolutional Neural Network
  • Architekturen
    • ImageNet-Architekturen: AlexNet, ZFNet, VGG-Net, InceptionNet, ResNet, ResNeXt, SENet
    • darauf aufbauende Architekturen: WideResNet, DenseNet, InceptionResNet, XceptionNet
    • Mobile Architekturen
    • Ausblick auf sonstige Architekturen
    • Kerntechnologien
      • ReLU und Erweiterungen
      • Dropout und Erweiterungen
      • Datenaugmentierung
      • Lernraten-Scheduling
      • Ersetzung von Convolutional Filtern mit großem rezeptiven Feld durch Stapel kleinerer Filter
      • Mikroarchitekturen
      • Bottleneck-Blöcke
      • Global Average Pooling
      • Batch Normalization und Erweiterungen
      • Residual-Block
      • Skip Connections
      • Grouped Convolutions, Depthwise Separable Convolutions
      • Squeeze-and-Excitation-Block
      • Stochastic Depth
  • Anwendungen
    • Detektion
    • Segmentierung
    • Posenerkennung
    • Wiedererkennung
    • Kerntechnologien
      • Region Proposal
      • Fully Convolutional Neural Network
      • moderne Fehlerfunktionen
  • Anwendung auf eigene Problemstellung
    • Umgang mit Daten
    • Auswahl geeigneter Architektur und Technik
    • Transfer Learning
    • Bewertungsmaße
    • Typische Probleme und abzuleitende Schlussfolgerungen (Best Practice Guide)
  • aktuelle Forschungsfragen

Vorlesungsunterlagen

 

Seminare

Angewandte Neuroinformatik

(Sommersemester 2014, 2015, 2016, 2017)

Zeit: Mo. (G) 17:00 Uhr

Inhalte

  • Module eines Mustererkennungsystems
  • Merkmalstransformation
  • Merkmalsselektion
  • Leistungsbewertung von Klassifikatoren
  • Ensemble Learning
  • Techniken zur Repräsentation von Zeit
  • Information Fusion

Vorlesungsunterlagen

Mensch-Maschine-Interaktion

(Wintersemester 2016/17)

Zeit: (Wintersemester)

Inhalte

  • Bildvorverarbeitung (Integralbild, Gaborfilter)
  • Hauptkomponentenanalyse (PCA)
  • Personendetektion (Eigenfaces, HOG)
  • Leistungsbewertung von Detektoren
  • Personentracking mittels Partikelfilter
  • Wiedererkennung von Personen
  • Hidden-Markov-Modell (HMM)

Vorlesungsunterlagen

Technische Informatik

(Wintersemester 2014/15, 2015/16)

Inhalte

  • Zahlensysteme
  • Grundrechenarten im Dualsystem
  • Logische Schaltungen
  • Boolsche Algebra
  • Darstellung negativer Zahlen
  • Zahlencodierungen (BCD, Aiken, 3XS)
  • Gleitkommazahlen
  • Assemblerbefehle
  • Turing-Maschine

Publications

Google Scholar Profile

 

2020

Mueller, S., Wengefeld, T., Trinh, T.Q., Aganian, D., Eisenbach, M., Gross, H.-M.
A Multi-Modal Person Perception Framework for Socially Interactive Mobile Service Robots.
Sensors, 20(3), 722, MDPI 2020
 

 

 

2019

Eisenbach, M.
Personenwiedererkennung mittels maschineller Lernverfahren für öffentliche Einsatzumgebungen.
Dissertation, TU Ilmenau, 2019
 

Eisenbach, M., Stricker, R., Sesselmann, M., Seichter, D., Gross, H.-M.
Enhancing the Quality of Visual Road Condition Assessment by Deep Learning.
in: World Road Congress 2019, Abu Dhabi, UAE, 13 pages, 2019
 

Schnürer, Th., Fuchs, St., Eisenbach, M.,  Gross, H.-M.
Real-Time 3D Pose Estimation from Single Depth Images.
in: Proc. Int. Joint Conf. on Computer Vision, Imaging and Computer Graphics - Theory and Applications (VISAPP), Prague, Czech Republic, pp. 716-724, 2019
 

Sesselmann, M., Stricker, R., Eisenbach, M.
Einsatz von Deep Learning zur automatischen Detektion und Klassifikation von Fahrbahnschäden aus mobilen LiDAR-Daten.
AGIT - Journal für Angewandte Geoinformatik, vol. 5-2019, pp. 100-114
Best Student Paper Award
 

Stricker, R., Eisenbach, M., Sesselmann, M., Debes, K., Gross, H.-M.
Improving Visual Road Condition Assessment by Extensive Experiments on the Extended GAPs Dataset.
in: Int. Joint Conf. on Neural Networks (IJCNN), Budapest, Hungary, paper N-20496, 8 pages, IEEE 2019
 

 

2018

Seichter, D., Eisenbach, M., Stricker, R., Gross, H.-M.
How to Improve Deep Learning based Pavement Distress Detection while Minimizing Human Effort.
in: Proc. Int. Conf. on Automation Science and Engineering (CASE), München, Germany, pp. 63-70, IEEE 2018
 

 

2017

Eisenbach, M., Stricker, R., Seichter, D., Amende, K., Debes, K., Sesselmann, M., Ebersbach, D., Stoeckert, U., Gross, H.-M.
How to Get Pavement Distress Detection Ready for Deep Learning? A Systematic Approach.
in: Proc. Int. Joint Conf. on Neural Networks (IJCNN), Anchorage, USA, pp. 2039-2047, IEEE 2017

Eisenbach, M., Stricker, R., Debes, K., Gross, H.-M.
Crack Detection with an Interactive and Adaptive Video Inspection System.
in: Arbeitsgruppentagung Infrastrukturmanagement, Duisburg, Germany, pp. 94-103, 2017
Eisenbach, M., Stricker, R., Seichter, D., Vorndran, A., Wengefeld, T., Gross, H.-M.
Speeding up Deep Neural Networks on the Jetson TX1.
in: IJCNN-Workshop on Computational Aspects of Pattern Recognition and Computer Vision with Neural Systems (CAPRI), Anchorage, USA, pp. 11-22, 2017
Gross, H.-M., Scheidig, A., Debes, K., Einhorn, E., Eisenbach, M., Mueller, S., Schmiedel, T., Trinh, T.Q., Weinrich, C., Wengefeld, T., Bley, A., Martin, C.
ROREAS - Robot Coach for Walking and Orientation Training in Clinical Post-Stroke Rehabilitation: Prototype Implementation and Evaluation in Fields Trials.
Autonomous Robots (AR), vol. 41 (2017) 3, pp. 679-698, Springer 2017

Gross, H.-M., Meyer, S., Scheidig, A., Eisenbach, M., Mueller, S., Trinh, T. Q., Wengefeld, T., Fricke, C.
Mobile Robot Companion for Walking Training of Stroke Patients in Clinical Post-stroke Rehabilitation.
in: Int. Conf. on Robotics and Automation (ICRA), Singapore, pp. 1028-1035, 2017
Best Paper Award on Human-Robot Interaction (HRI) - Finalist
 

 

2016

Eisenbach, M., Seichter, D., Wengefeld, T., Gross, H.-M.
Cooperative Multi-Scale Convolutional Neural Networks for Person Detection.
in: Proc. IEEE World Congress on Computational Intelligence (WCCI), Vancouver, Canada, pp. 267-276, IEEE 2016
Eisenbach, M., Seichter, D., Gross, H.-M.
Are Color Features Important for Person Detection? - Insights into Features Learned by Deep Convolutional Neural Networks.
in: Proc. 22nd German Color Workshop (FWS), Ilmenau, Germeny, pp. 169-182, 2016
Gross, H.-M., Eisenbach, M., Scheidig, A., Trinh, T.Q., Wengefeld, T.
Contribution towards Evaluating the Practicability of Socially Assistive Robots - by Example of a Mobile Walking Coach Robot.
to appear in: Proc. Int. Conf. on Social Robotics (ICSR), pp. 890-899, Springer 2016
Gross, H.-M., Scheidig, A., Eisenbach, M., Trinh, T.Q., Wengefeld, T.
Assistive Robotics for Health Assistance - a Contribution towards Evaluating the Practicability by Example of a Mobile Rehab Robot.
in: Proc. 9th German AAL Conference (AAL), Frankfurt, Germany, pp. 58-67, VDE Verlag 2016
Wengefeld, T., Eisenbach, M., Trinh, T.Q., Gross, H.-M.
May I be your Personal Coach? Bringing Together Person Tracking and Visual Re-identification on a Mobile Robot.
in: International Symposium on Robotics (ISR), Munich, Germany, pp. 141-148, 2015

 

2015

Eisenbach, M., Kolarow, A., Vorndran, A., Niebling, J., Gross, H.-M.
Evaluation of Multi Feature Fusion at Score-Level for Appearance-based Person Re-Identification.
in: Proc. Int. Joint Conf. on Neural Networks (IJCNN), Killarney, Ireland, pp. 469-476, IEEE 2015
Eisenbach, M., Vorndran, A., Sorge, S., Gross, H.-M.
User Recognition for Guiding and Following People with a Mobile Robot in a Clinical Environment.
in: Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), Hamburg, Germany, pp. 3600-3607, 2015
Scheidig, A.,  Einhorn, E., Weinrich, Ch.,  Eisenbach, M., Mueller, S., Schmiedel, T., Wengefeld, T., Trinh, Th., Gross, H.-M., Bley, A., Scheidig, R.,  Pfeiffer, G., Meyer, S., Oelkers, S.
Robotischer Reha-Assistent zum Lauftraining von Patienten nach Schlaganfall: Erste Ergebnisse zum Laufcoach.
in: Proc. 8th German AAL Conference (AAL), Frankfurt, Germany, pp. 436-445, VDE Verlag, 2015

 

2013

Eisenbach, M., Scheiner, P., Kolarow, A., Schenk, K., Gross, H.-M., Weinreich, I.
Learning Illumination Maps for Color Constancy in Person Reidentification.
in: Proc. 19th German Color Workshop (FWS), pp. 103-114, 2013

Kolarow, A., Schenk, K., Eisenbach, M., Dose, M., Brauckmann, M., Debes, K., Gross, H.-M.
APFel: The Intelligent Video Analysis and Surveillance System for Assisting Human Operators.
in: Proc. IEEE Int. Conf. on Advanced Video and Signal-Based Surveillance (AVSS), Krakow, Poland, pp. 195-201, IEEE 2013

 

2012

Eisenbach, M., Kolarow, A., Schenk, K., Debes, K., Gross, H.-M.
View Invariant Appearance-based Person Reidentification Using Fast Online Feature Selection and Score Level Fusion.
in: Proc. IEEE Int. Conf. on Advanced Video and Signal-Based Surveillance (AVSS), Beijing, China, pp. 184-190, IEEE 2012
Kolarow, A., Brauckmann, M., Eisenbach, M., Schenk, K., Einhorn, E., Debes, K., Gross, H.-M.
Vision-based Hyper-Real-Time Object Tracker for Robotic Applications.
in: Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), Vilamoura, Portugal, pp. 2108-2115, IEEE 2012
Schenk, K., Kolarow, A., Eisenbach, M., Debes, K., Gross, H.-M.
Automatic Calibration of a Stationary Network of Laser Range Finders by Matching Movement Trajectories.
in: Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), Vilamoura, Portugal, pp. 431-437, IEEE 2012
Schenk, K., Kolarow, A., Eisenbach, M., Debes, K., Gross, H.-M.
Automatic Calibration of Multiple Stationary Laser Range Finders using Trajectories.
in: Proc. IEEE Int. Conf. on Advanced Video and Signal-Based Surveillance (AVSS), Beijing, China, pp. 306-312, IEEE 2012

 

2011

Schenk, K., Eisenbach, M., Kolarow, A., Gross, H.-M.
Comparison of Laser-Based Person Tracking at Feet and Upper-Body Height.
in: Annual German Conf. on Artificial Intelligence (KI), Berlin, Germany, LNAI 7006, pp. 277–288,  Springer 2011

 

2009

Eisenbach, M.
Rewarddekomposition für Multiagentensysteme bei komplexen Regelungsprozessen.
Diplomarbeit, TU Ilmenau, 2009