Gesture Recognition in RGB Videos Using Human Body Keypoints and Dynamic Time Warping. Pascal Schneider, Raphael Memmesheimer, Ivanna Kramer, and Dietrich Paulus. In Stephan Chalup, Tim Niemueller, Jackrit Suthakorn, and Mary-Anne Williams, editors, RoboCup 2019: Robot World Cup XXIII, pages 281--293, Cham, 2019. Springer International Publishing. [ bib | DOI ]

Gesture recognition opens up new ways for humans to intuitively interact with machines. Especially for service robots, gestures can be a valuable addition to the means of communication to, for example, draw the robot’s attention to someone or something. Extracting a gesture from video data and classifying it is a challenging task and a variety of approaches have been proposed throughout the years. This paper presents a method for gesture recognition in RGB videos using OpenPose to extract the pose of a person and Dynamic Time Warping (DTW) in conjunction with One-Nearest- Neighbor (1NN) for time-series classification. The main features of this approach are the independence of any specific hardware and high flexibility, because new gestures can be added to the classifier by adding only a few examples of it. We utilize the robustness of the Deep Learning-based OpenPose framework while avoiding the data-intensive task of training a neural network ourselves. We demonstrate the classification performance of our method using a public dataset.

homer@UniKoblenz: Winning Team of the RoboCup@Home Open Platform League 2018. Raphael Memmesheimer, Ivanna Mykhalchyshyna, Viktor Seib, Tobias Evers, and Dietrich Paulus. In Dirk Holz, Katie Genter, Maarouf Saad, and Oscar von Stryk, editors, RoboCup 2018: Robot World Cup XXII [Montreal, QC, Canada, June 18-22, 2018]., volume XXII of Lecture Notes in Computer Science, pages 512--523. Springer International Publishing, 2019. [ bib | DOI ]

We won this year’s RoboCup@Home track in the Open Platform League in Montreal (Canada). The approaches as used for the competition are briefly described in this paper. The robotic hardware of our custom built robot Lisa and the PAL Robotics TIAGo, both running the same methods, are presented. New approaches for object recognition, especially the preprocessed segment augmentation, effort based gripping, gesture recognition and approaches for visual imitation learning based on continuous spatial observations between a demonstrator and the interacting objects are presented. Further, we present the current state of research of our Imitation Learning approaches, where we propose a hybrid benchmark and methods for bootstrapping actions. Furthermore, our research on point cloud based object recognition is presented.

homer@UniKoblenz: Winning Team of the RoboCup@Home Open Platform League 2017. Raphael Memmesheimer, Vikor Seib, and Dietrich Paulus. In Hidehisa Akiyama, Oliver Obst, Claude Sammut, and Flavio Tonidandel, editors, RoboCup 2017: Robot World Cup XXI, pages 509--520, Cham, 2018. Springer International Publishing. [ bib ]

In this paper we present the approaches that we used for this year's RoboCup@Home participation in the Open Platform League. A special focus was put on team collaboration by handing over objects between two robots of different teams that were not connected by network. The robots communicated using natural language (speech synthesis, speech recognition), a typical human-robot “interface” that was adapted to robot-robot interaction. Furthermore, we integrated new approaches for online tracking and learning of an operator, have shown a novel approach for teaching a robot new commands by describing them using natural language and a set of previously known commands. Parameters of these commands are still interchangeable. Finally, we integrated deep neural networks for person detection and recognition, human pose estimation, gender classification and object recognition.

RoboCup 2017 - homer@UniKoblenz (Germany), Raphael Memmesheimer, Viktor Seib, Niklas Yann Wettengel, Florian Polster, Daniel Müller, Moritz Löhne, Malte Roosen, Ivanna Mykhalchyshyna, Lukas Buchhold, Matthias Schnorr, and Dietrich Paulus. 2018. Technical Report. [ bib ]

This paper describes the robot Lisa used by team homer- @UniKoblenz of the University of Koblenz-Landau, Germany, for the participation at the RoboCup@Home 2017 in Nagoya, Japan. A spe- cial focus is put on novel system components and the open source con- tributions of our team. We have released packages for object recogni- tion, a robot face including speech synthesis, mapping and navigation, speech recognition interface via android and a GUI. The packages are available (and new packages will be released) on agas-ros-pkg.

RoboCup@Home: Summarizing achievements in over eleven years of competition. Mauricio Matamoros, Viktor Seib, Raphael Memmesheimer, and Dietrich Paulus. In 2018 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), pages 186--191, New York, 2018. IEEE. [ bib | DOI ]

Scientific competitions are important in robotics because they foster knowledge exchange and allow teams to test their research in unstandardized scenarios and compare result. In the field of service robotics its role becomes crucial. Competitions like RoboCup@Home bring robots to people, a fundamental step to integrate them into society. In this paper we summarize and discuss the differences between the achievements claimed by teams in their team description papers, and the results observed during the competition1 from a qualitative perspective. We conclude with a set of important challenges to be conquered first in order to take robots to people's homes. We believe that competitions are also an excellent opportunity to collect data of direct and unbiased interactions for further research.

Joint Operator Detection and Tracking for Person Following from Mobile Platforms. Nicolai Wojke, Raphael Memmesheimer, and Dietrich Paulus. In International Conference on Information Fusion (Fusion),, pages 905--912, 2017. [ bib | DOI ]

European Robotics League homer@UniKoblenz, Raphael Memmesheimer, Lukas Buchhold, and Florian Polster. 2017. [ bib ]

This paper describes the robot hardware and soft- ware used by team homer@UniKoblenz of the University of Koblenz and Landau, Germany, for the participation in the European Robotics League 2017 for Service Robots in Lisbon. A special focus is put on detailed hardware and software description and novel scientific achievements. For navigation and mapping we use well-established SLAM techniques based on particle filter and grid maps. Object recognition is achieved by clustering of local invariant features. Three-dimensional scans of the environment are acquired using a RGB-D sensor mounted on a pan-tilt unit. In order to manipulate objects a 6 DOF industrial-grade robotic arm is employed. An abstraction layer in our software architec- ture allows for task planning and general purpose task execution. For the improvement of human-robot interaction we developed a generic face model that is synchronized to speech and can show seven different face expressions. This robot face is available as a ROS-node for other teams. Some of the novelties of this year is a new robotic platform and a complete change to ROS that now replaces our previously used software architecture. Further, our object recognition, mapping and navigation algorithms were published as ROS package. Furthermore, improvements were made in speech recognition, object recognition, manipulation as well as 3D mapping.

Technical Report Team homer@UniKoblenz for ICRA / DJI RoboMasters, Raphael Memmesheimer, Florian Polster, Ivanna Mykhalchyshyna, Malte Roosen, and Lukas Buchhold. 2017. [ bib ]

In this document we give a technical report to show the progress of our team homer@UniKoblenz in the process of qualification. The corresponding report video is located on YouTube 1 where our robot “Grimey” stacks cubes autonomously up to a height of 80cm.

Friend or Foe: Exploiting Sensor Failures for Transparent Object Localization and Classification. Viktor Seib, Andreas Barthen, Philipp Marohn, and Dietrich Paulus. In International Conference on Robotics and Machine Vision (ICRMV) 2016, volume 10253 of Proc.SPIE, pages 10253.11--10253.15, 2017. [ bib | DOI | .pdf ]

Evaluation of Approaches Combining 2D and 3D Data for Object Recognition Developed for the Mobile Robot Lisa. Viktor Seib, Florian Polster, and Dietrich Paulus. In International Conference on Robotics and Machine Vision (ICRMV) 2016, volume 10253, pages 10253.1--10253.5, 2017. [ bib | DOI ]

A ROS-based System for an Autonomous Service Robot. Viktor Seib, Raphael Memmesheimer, and Dietrich Paulus. In Anis Koubaa, editor, Robot Operating System (ROS): The Complete Reference (Volume 1), volume 625 of Studies in Computational Intelligence, pages 215--252. Springer, Heidelberg, 2016. [ bib | DOI ]

Detecting Fine-grained Sitting Affordances With Fuzzy Sets. Viktor Seib, Malte Knauf, and Dietrich Paulus. In Nadia Magnenat-Thalmann, Paul Richard, Lars Linsen, Alexandru Telea, Sebastiano Battiato, Francisco Imai, and José Braz, editors, Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, pages 289--298, Setúbal, 2016. SciTePress. [ bib | DOI ]

Team Homer@ UniKoblenzApproaches and Contributions to the RoboCup@ Home Competition. Viktor Seib, Raphael Memmesheimer, Stephan Manthe, Florian Polster, and Dietrich Paulus. RoboCup 2015: Robot World Cup XIX, 9513:83--94, 2016. [ bib ]

RoboCup 2016 - homer@UniKoblenz (Germany). Raphael Memmesheimer, Viktor Seib, Gregor Heuer, Patrik Schmidt, Darius Thies, Ivanna Mykhalchyshyna, Johannes Klöckner, Martin Schmitz, Niklas Yann Wettengel, Nils Geilen, Richard Schütz, Florian Polster, and Dietrich Paulus. Technical report, University of Koblenz-Landau, 2016. [ bib ]

Service-Roboter. Dietrich Paulus, Viktor Seib, and Johannes Pellenz. In Matthias Gouthier, editor, Kundenbindung durch kosteneffiziente Service Excellence : Strategien - Konzepte - Best-Practices: Konferenzband der 5. Excellence-in-Service-Konferenz EXIS, pages 107--121, Baden-Baden, 2016. Nomos Verlag. [ bib | DOI ]

Shape Retrieval of Low-Cost RGB-D Captures. Pedro B. Pascoal, Pedro Proença, Filipe Gaspar, Miguel Sales Dias, Alfredo Ferreira, Atsushi Tatsuma, Masaki Aono, K. Berker Logoglu, Sinan Kalkan, Alptekin Temizel, Bo Li, Henry Johan, Yijuan Lu, Viktor Seib, Norman Link, and Dietrich Paulus. In A. Ferreira, A. Giachetti, and D. Giorgi, editors, Eurographics Workshop on 3D Object Retrieval, pages 69--78. The Eurographics Association, 2016. [ bib | DOI ]

Detecting Fine-grained Affordances with an Anthropomorphic Agent Model. Viktor Seib, Nicolai Wojke, Malte Knauf, and Dietrich Paulus. In David Fleet, Tomas Pajdla, Bernt Schiele, and Tinne Tuytelaars, editors, Computer Vision - ECCV 2014 Workshops, volume 8926 of LNCS, pages 413--419. Springer International Publishing Switzerland, 2015. [ bib | DOI ]

SHREC'15: Range Scans based 3D Shape Retrieval. Afzal Godil, H. Dutagaci, B. Bustos, S. Choi, S. Dong, T. Furuya, H. Li, Norman Link, A. Moriyama, R. Meruane, Ryutarou Ohbuchi, Dietrich Paulus, Tobias Schreck, Viktor Seib, Ivan Sipiran, H. Yin, and C. Zhang. In I. Pratikakis, M. Spagnuolo, T. Theoharis, Luc Van Gool, and Remco Veltkamp, editors, Eurographics Workshop on 3D Object Retrieval (2015), pages 153--160. The Eurographics Association, 2015. [ bib | DOI ]

Fourier Features For Person Detection in Depth Data. Viktor Seib, Guido Schmidt, Michael Kusenbach, and Dietrich Paulus. In George Azzopardi and Nicolai Petkov, editors, Computer Analysis of Images and Patterns (CAIP 2015), volume 9256 of LNCS, pages 824--836. Springer International Publishing Switzerland, 2015. [ bib | DOI ]

RoboCup 2015 - homer@UniKoblenz (Germany). Viktor Seib, Stephan Manthe, Jana Holzmann, Raphael Memmesheimer, Arne Peters, Markus Bonse, Florian Polster, Baharak Rezvan, Katrin Riewe, Malte Roosen, Uzaira Shah, Timur Yigit, Andreas Barthen, Malte Knauf, and Dietrich Paulus. Technical report, University of Koblenz-Landau, 2015. [ bib ]

homer@UniKoblenz. Viktor Seib, Raphael Memmesheimer, Florian Polster, Ivanna Mykhalchyshna, and Dietrich Paulus. Technical report, University of Koblenz-Landau, 2015. [ bib ]

homer@UniKoblenz. Viktor Seib, Nicolai Wojke, Raphael Memmesheimer, Daniel Müller, Sebastian Stümper, Florian Sattler, Alexander Kreutz, Sebastian Brockmann, Alruna Veith, Arne Peters, Denise Dünnebier, Eva Kreckel, Ivanna Mykhalchyshyna, and Dietrich Paulus. Technical report, University of Koblenz-Landau, 2014. [ bib ]

Object Recognition Using Hough-transform Clustering of SURF Features. Viktor Seib, Michael Kusenbach, Susanne Thierfelder, and Dietrich Paulus. In International Workshops on Electrical and Computer Engineering Subfields, pages 176--183. Scientific Cooperations Publications, 2014. [ bib ]

Enhancing Human-Robot Interaction by a Robot Face with Facial Expressions and Synchronized Lip Movements. Viktor Seib, Julian Giesen, Dominik Grüntjens, and Dietrich Paulus. In Vaclav Skala, editor, 21st International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG 2013), pages 70--77, Plzen, 2013. Union Agency. [ bib ]

RoboCup 2014 - homer@UniKoblenz (Germany). Viktor Seib, Raphael Memmesheimer, Tatjana Jakowlewa, David Nagel, Niklas Gard, Denise Dünnebier, Anatoli Eckert, Eva Kreckel, Alexander Kreutz, Ivanna Mykhalchyshyna, Greta Rettler, Florian Sattler, Alruna Veith, Malte Knauf, Arne Peters, and Dietrich Paulus. Technical report, University of Koblenz-Landau, 2014. [ bib ]

RoboCup 2013 - homer@UniKoblenz (Germany). Viktor Seib, Florian Kathe, Daniel McStay, Stephan Manthe, Arne Peters, Benedikt Jöbgen, Raphael Memmesheimer, Tatjana Jakowlewa, Caroline Vieweg, Sebastian Stümper, Sebastian Günther, Simon Müller, Alruna Veith, Michael Kusenbach, Malte Knauf, and Dietrich Paulus. Technical report, Universität Koblenz-Landau,, 2013. [ bib ]

RoboCup 2012 - homer@UniKoblenz (Germany). Viktor Seib, Veronica Luing, Lubosz Sarnecki, Thomas Grün, Malte Knauf, Andreas Barthen, Liane Syré, Alruna Veith, Olaf Poneta, Sonja Polster, Michael Kusenbach, Marco Maas, David Nagel, Julian Giesen, Susanne Thierfelder, and Dietrich Paulus. Technical report, Universität Koblenz-Landau,, 2012. [ bib ]

Object Recognition Tasks for Service Robots. Viktor Seib, Susanne Thierfelder, and Dietrich Paulus. In 8th Open German-Russian Workshop: Pattern Recognition and Image Understanding (OGRW-8-2011), pages 258--261, 2011. [ bib ]

Hierarchical Multi-robot Coordination. Viktor Seib, David Gossow, Sebastian Vetter, and Dietrich Paulus. In Javier Ruiz-del Solar, Eric Chown, and Paul G. Plöger, editors, RoboCup 2010: Robot Soccer World Cup XIV, volume 6556 of LNCS, pages 314--323, Berlin, 2011. Springer. [ bib | DOI ]

RoboCup 2011 - homer@UniKoblenz (Germany). Susanne Thierfelder, David Gossow, Carmen Navarro Luzón, Sebastian Nowack, Nico Merten, Susanne Friedmann, Lydia Rebecca Weiland, Daniel Mies, Julian Giesen, Marcel Häselich, Dagmar Lang, Viktor Seib, and Dietrich Paulus. Technical report, Universität Koblenz-Landau,, 2011. [ bib ]

Robbie: A Message-based Robot Architecture for Autonomous Mobile Systems. Susanne Thierfelder, Viktor Seib, Dagmar Lang, Marcel Häselich, Johannes Pellenz, and Dietrich Paulus. In Hans-Ulrich Heiß, Peter Pepper, Holger Schlingloff, and Jörg Schneider, editors, INFORMATIK 2011 - Informatik schafft Communities, volume 191 of Lecture Notes in Informatics (LNI), page 331. Köllen Druck+Verlag GmbH Bonn, 2011. CD-ROM. [ bib ]

RoboCup 2009 - homer@UniKoblenz (Germany). David Gossow, Marc Arends, Susanne Thierfelder, Christian Winkens, Sönke Greve, Johannes Pellenz, Peter Decker, Viktor Seib, and Dietrich Paulus. Technical report, Universität Koblenz-Landau, http://www.uni-, 2009. [ bib ]