My name is Marco Peer and I joined the CVL as a university assistent in 2021. My main research interests are in the field of document analysis and content-based image retrieval with deep learning. My PhD is focused on writer retrieval for historical documents.

I also worked in the WRITE project, in which I developed writer retrieval methods to identify suspects for the Bundeskriminalamt.

Teaching

• Summer term
  • Einführung in Visual Computing (VU)
  • Document Analysis (VU)
• Winter term
  • Seminar aus Computer Vision (und Mustererkennung) (SE)
  • Wissenschaftliches Arbeiten (SE)
  • Computer Vision (VU)
• Supervision of Bachelor and Master theses as well as student projects.

Publications 

• M. Muth, M. Peer, F. Kleber and R. Sablatnig, “Advancing Handwritten Text Detection by Synthetic Text”,  accepted for ICPR2024, Kolkata, India, December 2024.
• M. Muth, M. Peer, F. Kleber and R. Sablatnig, “Maximizing Data Efficiency of HTR Models by Synthetic Text”,  accepted for DAS@ICDAR2024, Athens, Greece, August 2024.
• M. Peer, R. Sablatnig, O. Serbaeva and I. Marthot-Santaniello, “KaiRacters: Character-level-based Writer Retrieval for Greek Papyri”, submitted to ICPR2024. arXiv, Dataset
• M. Peer, F. Kleber and R. Sablatnig, “SAGHOG: Self-Supervised Autoencoder for Generating HOG Features for Writer Retrieval”, accepted to ICDAR2024, Athens, Greece, September 2024. arXiv, Code
• V. Pundy, M. Peer, F. Kleber, “Transparency Techniques for Neural Networks trained on Writer Identification and Writer Verification”, presented at Applied Vision@AIRoV – The First Austrian Symposium on AI, Robotics, and Vision, Innsbruck, Austria, February 2024.
• M. Peer and R. Sablatnig, “Feature Mixing for Writer Retrieval and Identification on Papyri Fragments”, in Proceedings of the 7th International Workshop on Historical Document Imaging and Processing, HIP@ICDAR 2023, San José, California, USA, August 2023. (oral presentation) arXiv , Code
• M. Peer, F. Kleber and R. Sablatnig, “Towards Writer Retrieval for Historical Datasets,” in Document Analysis and Recognition – ICDAR 2023 – 17th International Conference, San José, California, USA, August 2023. (oral presentation) arXiv , Code
  
• G. Heilemann, L. Zimmermann, R. Schotola, W. Lechner, M. Peer et al., “Generating deliverable DICOM RT treatment plans for prostate VMAT by predicting MLC motion sequences with an encoder-decoder network”, Med Phys. 2023; 1– 7. https://doi.org/10.1002/mp.16545
• M. Peer, F. Kleber and R. Sablatnig,  “Self-supervised Vision Transformers with Data Augmentation Strategies Using Morphological Operations for Writer Retrieval,” in Frontiers in Handwriting Recognition, Hyderabad, India, December 2022 pp. 122–136. (oral presentation)
• M. Peer, F. Kleber and R. Sablatnig, “Writer Retrieval using Compact Convolutional Transformers and NetMVLAD,” in 2022 26th International Conference on Pattern Recognition (ICPR), Montreal, QC, Canada, August 2022 pp. 1571-1578.
• M. Peer,  S. Thalhammer and M. Vincze, “Image Synthesis in SO(3) by Learning Equivariant Features Spaces,” in Proceedings of the Joint Austrian Computer Vision and Robotics Workshop 2020, 2020, p. 108–113.

Presentations

• C. Li, M. Peer and K. Yang, “Looking for Chos grub: Proposing An AI-Assisted Method for Recognizing Scribal Hands of Dunhuang Tibetan Manuscripts”, Advanced Computational Methods for
  Studying Buddhist Texts. Vienna, Austria. April 27, 2023.
• M. Peer, “WRITE – IT unterstützte Suche und Vergleich von Handschriften”, 12. KIRAS Fachtagung. Vienna, Austria. October 25, 2022.

Awards

• Best Paper Nomination at AIRoV 2024 for Transparency Techniques for Neural Networks trained on Writer Identification and Writer Verification.
• Best Poster Award: “Writer Retrieval for Historical Documents”. The 5th IAPR TC10/TC11 Summer School on Document Analysis. Fribourg, Switzerland. 2023.
• Excellence Award. The 5th IAPR TC10/TC11 Summer School on Document Analysis. Fribourg, Switzerland. 2023.
• Best Contribution Award: My diploma thesis was part of the study „Generating treatment plans from dose distributions on segmented CTs with Deep Learning: A feasibility study“ awarded by the Österreichische Gesellschaft für Medizinische Physik (ÖGMP).

Education

2021	Dipl.-Ing. TU Wien. Energy Systems and Control Engineering. Thesis title: Generating Treatment Plans for Prostate VMAT using Deep Learning. Advisor: Georg Dietmar
2018	BSc. TU Wien. Electrical Engineering. Thesis title: Modellierung eines Langstator Linearmotors. Advisor: Andreas Kugi
2015	Secondary School BG/BRG Waidhofen an der Thaya