Mitarbeiter IES

M.Sc. Andreas Specker

  • Karlsruher Institut für Technologie – KIT
    Lehrstuhl für Interaktive Echtzeitsysteme (IES)
    Prof. Dr.- Ing. Jürgen Beyerer
    c/o Technologiefabrik
    Haid-und-Neu-Str. 7
    76131 Karlsruhe

Lebenslauf

Andreas Specker absolvierte sein Bachelorstudium der Technischen Informatik von 2012 bis 2015 an der Dualen Hochschule Baden-Württemberg in Heidenheim und bei der Robert Bosch Automotive Steering GmbH. Anschließend studierte er von 2015 bis 2018 Informatik am Karlsruher Institut für Technologie. Seine Masterarbeit zum Thema „Leveraging Semantic Attributes and Human-in-the-loop for Practical Deep Learning-based Person Re-identification“ verfasste er in der Abteilung Videoauswertesysteme (VID) am Fraunhofer IOSB und am Lehrstuhl für Interaktive Echtzeitsysteme (IES). Seit Ende 2018 ist Herr Specker als wissenschaftlicher Mitarbeiter an diesem Lehrstuhl angestellt und arbeitet weiterhin in enger Kooperation mit der Abteilung VID des Fraunhofer IOSB zusammen. Aufbauend auf seiner Masterarbeit beschäftigt er sich derzeit mit Methoden zur attributbasierten Personenwiedererkennung mithilfe von Deep Learning-Verfahren.

Veröffentlichungen


2021
Deep learning-based video analysis pipeline for person detection and re-identification in aerial imagery.
Specker, A.; Moritz, L.; Sommer, L. W.
2021. Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies V, 118690K, SPIE. doi:10.1117/12.2600109
Improving Attribute-Based Person Retrieval By Using A Calibrated, Weighted, And Distribution-Based Distance Metric.
Specker, A.; Beyerer, J.
2021. 2021 IEEE International Conference on Image Processing (ICIP), Anchorage, AK, USA, Anchorage, AK, USA, 19-22 Sept. 2021, 2378–2382, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/ICIP42928.2021.9506096
Deep learning based person search in aerial imagery.
Sommer, L.; Specker, A.; Schumann, A.
2021. SPIE DEFENSE + COMMERCIAL SENSING | 12-17 APRIL 2021 - Automatic Target Recognition XXXI. Ed.: R. I. Hammoud, Art.-Nr.: 117290O, Society of Photo-optical Instrumentation Engineers (SPIE). doi:10.1117/12.2588179
A study of person re-identification design characteristics for aerial data.
Moritz, L.; Specker, A.; Schumann, A.
2021. Pattern Recognition and Tracking XXXII: SPIE DEFENSE + COMMERCIAL SENSING | 12-17 APRIL 2021. Ed.: M. S. Alam, Art.-Nr.: 117350P, Society of Photo-optical Instrumentation Engineers (SPIE). doi:10.1117/12.2587946
A Step Towards Explainable Person Re-identification Rankings.
Specker, A.
2021. Proceedings of the 2020 Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory. Ed.: J. Beyerer; T. Zander, 107–121, KIT Scientific Publishing
An Occlusion-Aware Multi-Target Multi-Camera Tracking System.
Specker, A.; Stadler, D.; Florin, L.; Beyerer, J.
2021. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 4168–4177, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/CVPRW53098.2021.00471
2020
A multitask model for person re-identification and attribute recognition using semantic regions.
Specker, A.; Schumann, A.; Beyerer, J.
2020. Artificial Intelligence and Machine Learning in Defense Applications II. Ed.: J. Dijk, 115430I, SPIE. doi:10.1117/12.2573981
Large Scale Vehicle Re-Identification by Knowledge Transfer From Simulated Data and Temporal Attention.
Eckstein, V.; Schumann, A.; Specker, A.
2020. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020, Virtual, Online, United States, 14-19 June 2020, 2626–2631, IEEE Computer Society. doi:10.1109/CVPRW50498.2020.00316
The MTA Dataset for Multi-Target Multi-Camera Pedestrian Tracking by Weighted Distance Aggregation.
Kohl, P.; Specker, A.; Schumann, A.; Beyerer, J.
2020. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020, Virtual, Online, United States, 14 - 19 June 2020, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/CVPRW50498.2020.00529
A Realistic Predictor for Pedestrian Attribute Recognition.
Specker, A.
2020. Proceedings of the 2019 Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory. Hrsg.: J. Beyerer; T. Zander, 95–112, KIT Scientific Publishing
2019
An Interactive Framework for Cross-modal Attribute-based Person Retrieval.
Specker, A.; Schumann, A.; Beyerer, J.
2019. 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), 2019, Taipei, Taiwan, 18-21 Sept. 2019, 8909832, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/AVSS.2019.8909832
2018
Attribute-based Person Retrieval and Search in Video Sequences.
Schumann, A.; Specker, A.; Beyerer, J.
2018. 2018 15th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), 1–6, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/AVSS.2018.8639114