Ksenia Bittner (née Davydova)
Research Scientist at German Aerospace Center (DLR), Machine Learning Specialist
I have recently completed my Ph.D. degree under the guidance of Prof. Dr.-Ing. Peter Reinartz and Dr. rer. nat. Marco Körner at the German Aerospace Center (DLR) and successfully graduated from Osnabrück University in Mathematics and Informatics. I obtained my M.Sc. degree in Earth Oriented Space Science and Technology - an international Master's Program at the Technical University of Munich for Space Engineering and its applications: Earth System Science, Remote Sensing, and Navigation.
Currently, I am working as a Research Scientist at the German Aerospace Center (DLR).
My interest lies broadly in the area of machine learning, deep learning, artificial intelligence (AI), and its application to photogrammetry and remote sensing problems. More specifically, I am working on 3D buildings reconstruction from Digital Surface Models using deep learning techniques.
Education
Osnabrück University, Lower Saxony, Osnabrück, Germany
Majors/emphasizes: Geoinformatics and Machine Learning
PhD Thesis: Building Information Extraction and Refinement from VHR Satellite Imagery using Deep Learning Techniques
Thesis grade: summa cum laude
2012-2015 Master's Degree, Earth Oriented Space Science and Technology
Technische Universität München (TUM), Munich, Germany
Majors/emphasizes: Photogrammetry and Remote Sensing
Master Thesis: Consistent Multi-View Texturing of detailed 3D Surface
Models of Urban Areas Thesis grade: 1.3 ("very good")
National Research University MEPhI (Moscow Engineering Physics Institute), Moscow, Russia
Majors/emphasizes: Electronics and Automatics of Nuclear Power Plants
Diploma Thesis: Development of equipment for testing devices of automation
radiation monitoring systems of nuclear power plants (by using MODBUS protocol)
Thesis grade: 1.0 ("very good")
Work Experience
09/2019-present Research scientist/ Machine Learning Specialist
German Aerospace Center (DLR), Oberpfaffenhofen, Germany
Main tasks: Machine learning algorithms development for diverse photogrammetry and remote sensing problems.
09/2015-09/2019 Research assistant/PhD candidate
German Aerospace Center (DLR), Oberpfaffenhofen, Germany
Main tasks: Integration of multiple high-resolution remote sensing data sources for automatic satellite imagery interpretation with emphasis on building information extraction and refinement applying deep learning techniques.
05/2014-11/2014 Master student at Photogrammetry and Image Analysis Department
German Aerospace Center
(DLR), Oberpfaffenhofen, Germany
Main tasks: Creation of a "seamless" texturing of a 3D scene, using as input
a triangulated mesh of the scene and multiple camera views from different directions
German Aerospace Center
(DLR) Oberpfaffenhofen, Germany
06/2013-12/2013 Student assistant at Earth Oriented Space Science and Technology Technische Universität München, Munich, Germany
Main tasks: Students assistance, event organization
03/2011-08/2012 Engineer at Innovation Technologies DepartmentScientific Research Institute for Instrumentation Engineering, Moscow, Russia
Main tasks: Took part in development of equipment for testing devices of automation radiation monitoring systems of nuclear power plants
09/2009-06/2010 Trainee at Automation Systems DepartmentRussian Research Institute of Automatics, Moscow, Russia
Main tasks: Took part in development of multiplex channel debug modulebased on programmable logic device (PLD) Altera
Publications
- D'Angelo, Pablo and Cerra, Daniele, and Azimi, Seyed Majid and Merkle, Nina and Tian, Jiaojiao and Auer, Stefan and Pato, Miguel and Reyes, Raquel and Zhuo, Xiangyu and Bittner, Ksenia and Krauss, Thomas and Reinartz, Peter: 3D Semantic Segmentation from Multi-View Optical Satellite Images. In: IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2019. S. 5053-5056.
- Bittner, Ksenia and Körner, Marco and Reinartz, Peter. DSM Building Shape Refinement from Combined Remote Sensing Images based on WNet-cGANs. arXiv preprint arXiv:1903.03519, 2019.
- Bittner, Ksenia and Körner, Marco and Reinartz, Peter. Late or Earlier Information Fusion from Depth and Spectral Data? Large-Scale Digital Surface Model Refinement by Hybrid-cGAN. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2019.
- Schuegraf, Philipp and Bittner, Ksenia. Automatic Building Footprint Extraction from Multi-Resolution Remote Sensing Images Using a Hybrid FCN. ISPRS International Journal of Geo-Information, 2019, 8. Jg., Nr. 4, S. 191.
- Bittner, Ksenia and Körner, Marco and Fraundorfer, Friedrich and Reinartz, Peter:
Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-Type Classification. Remote Sensing, 2019, 11. Jg., Nr. 11, S. 1262.
- Cerra, Daniele and Pato, Miguel and Carmona, Emiliano and Azimi, Seyed Majid and Tian, Jiaojiao and Bahmanyar, Reza and Kurz, Franz and Vig,Eleonora and Bittner, Ksenia and Henry, Corentin and d'Angelo, Pablo and Müller, Rupert and Alonso, Kevin and Fischer, Peter and Reinartz, Peter: Combining Deep and Shallow Neural Networks with Ad Hoc Detectors for the Classification of Complex Multi-Modal Urban Scenes
- Bittner, Ksenia and dAngelo, Pablo and Körner, Marco and Reinartz, Peter: DSM-to-LoD2: Spaceborne Stereo Digital Surface Model Refinement. Remote Sensing, 2018, 10. Jg., Nr. 12, S. 1926.
- Bittner, Ksenia and Adam, Fathalrahman and Cui, Shiyong and Körner, Marco and Reinartz, Peter Peer-reviewed accepted paper: Building Footprint Extraction from VHR Remote Sensing Images Combined with Normalized DSMs using Fused Fully Convolutional Networks. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11. Jg., Nr. 8, S. 2615-2629.
- Bittner, Ksenia and Marco Korner: Automatic Large-Scale 3D Building Shape Refinement Using Conditional Generative Adversarial Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2018.
- Bittner, Ksenia and d'Angelo, Pablo and Körner, Marco
and Reinartz, Peter (2018) Automatic Large-Scale 3D Building
Shape Refinement Using Conditional Generative Adversarial Networks. International Archives of the Photogrammetry, Remote Sensing & Spatial Information Sciences 42.2 (2018). (Oral)
- Bittner, Ksenia and Cui, Shiyong and Reinartz, Peter (2017) Building extraction from remote sensing data using fully convolutional networks.
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-1/W1, 2017 ISPRS Hannover Workshop: HRIGI 17 - CMRT 17 - ISA 17 - EuroCOW 17, 6-9 June 2017, Hannover, Germany (Poster) Davydova, Ksenia and Cui, Shiyong and Reinartz, Peter (2016) Building footprint extraction from Digital Surface Models using Neural Networks. In: Proceedings of SPIE, 10004, pp. 1-10. SPIE Remote Sensing 2016, 26-29 Sep 2016, Edinburgh. DOI: 10.1117/12.2240727 (Oral)
Davydova, Ksenia and Kuschk, Georg and Hoegner, Ludwig and Reinartz, Peter and Stilla, Uwe (2015) Consistent Multi-View Texturing of Detailed 3D Surface Models. In: ISPRS Annals of the Photogrammetry Remote Sensing and Spatial Information Sciences, pp. 25-31. Copernicus Publications. Photogrammetric Image Analysis (PIA) 2015, 25.-27. March 2015, München, Deutschland. DOI: 10.5194/isprsannals-II-3-W4-25-2015 (Oral)
Awards
Asto Aerospace Promotion Prize (2017)