Vincent Christlein

Dr.-Ing. Vincent Christlein

Researcher

Department of Computer Science
Chair of Computer Science 5 (Pattern Recognition)

Room: Room 09.135
Martensstraße 3
91058 Erlangen

http://orcid.org/0000-0003-0455-3799

News:

Vincent Christlein studied Computer Science, graduated in 2012 and received his PhD (Dr.-Ing.) in 2018 from the Friedrich-Alexander University of Erlangen-Nürnberg (FAU), Germany. During his studies, he worked on automatic handwriting analysis with focus on writer identification and writer retrieval. Since 2018, he has been working as a research associate at the Pattern Recognition Lab (FAU) and was promoted to Academic Councilor in 2020. He heads the Computer Vision group, which covers a wide variance of topics, e.g. environmental projects such as glacier segmentation or solar cell crack recognition, but also computational humanities topics, such as document and art analysis.

2021

Best Student Paper Award for the paper:
Mattick, A.; Mayr, M.; Seuret, M.; Maier, A.; Christlein, V.: “SmartPatch: Improving Handwritten Word Imitation iwth Patch Discriminators”,
16th International Conference on Document Analysis and Recognition (ICDAR), 2021

 

2020

Best Student Paper Award (Honorable Mention) for the paper:
Leipert, M.; Vogeler, G.; Seuret M.; Maier A.; Christlein, V.: “The Notary in the Haystack – Countering Class Imbalance in Document Processing with CNNs”,
14th IAPR workshop on Document Analysis Systems (DAS), 2020.

 

2018
Best Student Paper Award for the paper:
Stromer D., Christlein V., Huang X., Zippert P., Helmecke E., Hausotte T., Maier A.:
“Non-Destructive Digitization of Soiled Historical Chinese Bamboo Scrolls”,
13th IAPR Workshop on Document Analysis Systems (DAS), 2018

 

2017

IEEE SPS Award: on behalf of IEEE Signal Processing Society (SPS) for the paper:
Christlein, Vincent; Riess, Christian; Jordan, Johannes; Riess, Corinna; Angelopoulou, Elli: “An Evaluation of Popular Copy-Move Forgery Detection Approaches”, IEEE Transactions on Information Forensics and Security, vol. 7, no. 6, pp. 1841-1854, 2012
The SPS Best Paper Award is regularly given to high impact journal papers in the last five years and is based on a nomination from the respective IEEE Technical Committee.
Competition Winner: ICDAR2017 Competition on the Classification of Medieval Handwritings in Latin Script: Task 2 (Style classification on heterogeneously encoded data) and Task 4 (Manuscript dating on heterogeneously encoded data)

 

2016

Competition Winner: ICFHR2016 Competition on Multi-Script Writer Demographics Classification Using “QUWI” Database: Task 1 (Writer Identification) and Task 3 (Handedness Classification)
Competition Winner: ICFHR2016 Competition on the Classification of Medieval Handwritings in Latin Script: Task 1 (Accuracy)

 

2015

Best Poster Award for a very significant contribution in the field of document analysis and recognition (ICDAR 2015)
Competition Winner: ICDAR 2015 Competition on Multi-Script Writer Identification and Gender Classification (Task 1 Writer identification)

 

2014

Competition Winner: ICFHR2014 Competition on Arabic Writer Identification Using AHTID/MW and KHATT Databases (Task 3 Writer identification using the KHATT datasbase)

2021

  • Font Group Recognition for Improved OCR

    (Third Party Funds Single)

    Term: August 1, 2021 - August 1, 2023
    Funding source: DFG-Einzelförderung / Sachbeihilfe (EIN-SBH)

    Although OCR-D made huge progress in the last project phase in providing OCR for early printed books, it still faces two major problems: The huge variety of the material makes it extremely challenging to use generic OCR-models. Yet, selecting specific models is not possible as the sheer amount of material prevents a fully automatic workflow. This situation is further complicated by the lack of appropriate OCR training data. Current data sets consist overwhelmingly of texts in Fraktur, especially from the 19th century. This completely neglects the large typographic variety displayed by printing in the three previous centuries. Therefore, and in response to the demand from SLUB Dresden and ULB Halle, we propose to improve the current situation significantly1) fine tuning our font group recognition system to such a degree that it can be used at character level;2) transcribing more specific OCR training data for the 16th-18th century, which includes popular fonts such as Schwabacher, other bastards and old Fraktur styles; 3) training font-specific OCR models as well as integrated models that recognise both typeface and text simultaneously. This approach has ensured in other contexts that the network performs better on both individual tasks, as we can thus reduce overfitting during training. This project will improve OCR quality significantly, especially for books in non-Fraktur fonts. It will also provide a training data set of very high quality that can be reused in long term. Finally, the project will provide a more fine-grained font recognition tool that, beyond enabling font-specific OCR, also has important applications in text attribute recognition and layout analysis.

2019

  • Kommunikation und Sprache im Reich. Die Nürnberger Briefbücher im 15. Jahrhundert: Automatische Handschriftenerkennung - historische und sprachwissenschaftliche Analyse.

    (Third Party Funds Single)

    Term: October 1, 2019 - September 30, 2022
    Funding source: DFG-Einzelförderung / Sachbeihilfe (EIN-SBH)

2018

  • Critical Catalogue of Luther Portraits (1519-1530)

    (Third Party Funds Group – Overall project)

    Term: June 1, 2018 - May 31, 2021
    Funding source: andere Förderorganisation
    URL: https://www.gnm.de/forschung/projekte/luther-bildnisse/

    Quite a number of portraits of Martin Luther - known as media star of 16th century – can be found in today’s museums and libraries. However, how many portraits indeed exist and which of those are contemporary or actually date from a later period? So far unlike his works, however, the variety of contemporary portraits (painting and print) is neither completely collected nor critically analyzed. Thus, a joint project of the FAU, the Germanisches Nationalmuseum (GNM) in Nuremberg and the Technology Arts Sciences (TH Köln) was initiated. Goal of the interdisciplinary project covering art history, art technology, reformation history and computer science is the creation of a critical catalogue of Luther portraits (1519-1530). In particular, the issues of authenticity, dating of artworks and its historical usage context as well as the potential existence of serial production processes will be investigated.

  • Similarity learning for art analysis

    (Third Party Funds Group – Sub project)

    Overall project: Critical Catalogue of Luther Portraits (1519-1530)
    Term: June 1, 2018 - February 28, 2021
    Funding source: andere Förderorganisation
    URL: https://www.gnm.de/forschung/forschungsprojekte/luther-bildnisse/

    The analysis of the similarity of portraits is an important issue for many sciences such as art history or digital humanities, as for instance it might give hints concerning serial production processes, authenticity or temporal and contextual classification of the artworks.
    In the project, first algorithms will be developed for cross-genre and multi-modal registration of portraits to overlay digitized paintings and prints as well as paintings acquired with different imaging systems such as visual light photography and infrared reflectography. Then, methods will be developed to objectively analyze the portraits according to their similarity.
    This project is part of a joint project of the FAU, the Germanisches Nationalmuseum (GNM) in Nuremberg and the Technology Arts Sciences (TH Köln) in Cologne. Goal of the interdisciplinary project covering art history, art technology, reformation history and computer science is the creation of a critical catalogue of Luther portraits (1519-1530).

2022

Journal Articles

Conference Contributions

2021

Journal Articles

Conference Contributions

2020

Journal Articles

Book Contributions

Conference Contributions

2019

Journal Articles

Conference Contributions

Thesis

2018

Journal Articles

Book Contributions

Edited Volumes

Conference Contributions

2017

Journal Articles

Book Contributions

Conference Contributions

Miscellaneous

2016

Journal Articles

Conference Contributions

2015

Conference Contributions

2014

Conference Contributions

2013

Conference Contributions

2012

Journal Articles

2010

Conference Contributions

2022

Christlein, Vincent
Automatische (Textunabhängige) Schreiberidentifizierung
at Tagung der Gesellschaft für forensische Schriftuntersung, Königswinter (17.06.)

Christlein, Vincent
Maschinelles Lernen zur Analyse historischer Dokumente
at Lange Nacht der Wissenschaften, Erlangen (21.05.)

Christlein, Vincent
Machine Learning for the Analysis of Artworks and Historical Documents
at Digital Humanities and Artificial Intelligence (DHAI) Seminar (10.05.)

Christlein, Vincent; Mayr, Martin
Handwriting Generation
at Weihnachtskolloquium 2020/21, AG OCR (25.01.)

2021

Christlein, Vincent
Maschinelles Lernen zur Bilddatenanalyse von Historischen Dokumenten
at Ringvorlesung “Materialität in den Historischen Grundwissenschaften – Aktuelle Technologien, die Digital Humanities und neue Wege der Forschung”, Ludwig-Maximilians University Munich (17.11.)

Christlein, Vincent
Handwriting Analysis and Generation
at Lunchtime Lectures, University of Graz (30.06.)

Christlein, Vincent; Mayr, Martin
Handwritten Text Recognition (HTR) – Entwicklung und aktuelle Ansätze der Handschriftenerkennung
at Weihnachtskolloquium 2020/21, AG OCR (21.01.)

2020

Christlein, Vincent
Automatische Schreiberidentifizierung in historischen Dokumenten
at “Digital History”, Digitale Geschichtswissenschaften an der Humboldt-Universität zu Berlin (09.11.)

Christlein, Vincent; Stromer, Daniel
How to Read the Unreadable? Non-Invasive digitization of cultural heritage
at Dark Archives Conference 2020 (09.09.)

Christlein, Vincent; Weichselbaumer, Nikolaus
The rapid rise of Fraktur
at University of Paderborn (DHd2020) in Paderborn (05.03)

Christlein, Vincent
Writer Identification in Historical Document Images
at Universität Basel, Switzerland (Neo-Palaeography: Analysing Ancient Handwritings in the Digital Age) in Basel (21.01)

2019

Christlein, Vincent
Automatische Bilddatenanalyse von Historischen Dokumenten
at Mathematikon, University of Heidelberg (GG3D19: Graphdatenbanken, GIS und 3D-Modelle in der Bauforschung des Mittelalters) in Heidelberg (5.12.)

Christlein, Vincent
Deep Generalized Max-Pooling
at Sydney, Australia (International Conference on Document Analaysis and Recognition) in Sydney (25.09)

at Faculty of English, Oxford (Dark Archives: A Conference on the Medieval Unread and Unreadable) in Oxford (11.09)

Christlein, Vincent
Maschinelles Lernen zur bildbasierten Dokumentenanalyse (Talk)
at Digital Humanities (Digital Humanities Cologne Colloquium) in University of Cologne (09.05)

Tutorial: Deep Learning Fundamentals (Talk)
at Institut für Medizinische Informatik, Universität zu Lübeck (Workshop Bildverarbeitung für die Medizin) in Lübeck (17.03)
2018

Christlein, Vincent
Encoding CNN Activations for Writer Recognition (Talk)

at TU Wien (13th IAPR International Workshop on Document Analysis Systems (DAS)) in Wien (24.04)

Deep Learning: Fundamentals (Talk)
at Lehrstuhl für Mustererkennung (Bildverarbeitung für die Medizin) in Erlangen (11.03)
at Exzellenzcluster (GI-Workshop: Im Spannungsfeld zwischen Tool-Building und Forschung auf Augenhöhe – Informatik und die Digital Humanities) in Berlin (25.09)
Handwriting Analysis with Focus on Writer Recognition (Talk)
at Universität Hamburg (Invited Talk) in Informatikum Stellingen, Hamburg (26.10)
at Ecole polytechnique fédérale de Lausanne (Time Machine Conference 2018) in Lausanne (30.10)
Visuelles Suchen in historischen Dokumenten – Herausforderungen und Möglichkeiten (Visual Retrieval in Historical Documents – Challenges and Possibilities) (Talk)
at ZIP O.P.S. (Archivrecherchen im 21. Jahrhundert) in Stift Teplá (08.11)
Writer Recognition in Historical Documents (Talk)
at Stadtarchäologie Wien (Congress Visual Heritage 2018) in Rathaus, Wien (14.11)

 

2017

Christlein, Vincent
Classification of Medieval Handwritten Script (Talk)
at Interdisziplinäre Zentrum Editionswissenschaften (IZED), Friedrich-Alexander-Universität Erlangen-Nürnberg (Workshop Digitale Paläographie) in Erlangen (31.01)

Christlein, Vincent; Ghesu, Florin Cristian; Maier, Andreas; Würfl, Tobias; Isensee, Fabian; Maier-Hein, Klaus; Neher, Peter; Kohl, Simon
Deep Learning: Advancing the State-of-the-Art in Medical Image Analysis (Talk)
at Deutsches Krebsforschungszentrum in der Helmholtz-Gemeinschaft (Bildverarbeitung für die Medizin 2007) in Heidelberg (12.03)
Automatic Handwriting Analysis – From Scribe to Script Type Classification (Talk)
at Ecole Nationale des Chartes (Seminaire e-Philologie) in Paris (27.03)
at IAPR (14th IAPR International Conference on Document Analysis and Recognition (ICDAR)) in Kyoto, Japan (15.11)
2016

Christlein, Vincent
Writer Retrieval and its Application to Music Scores
at Bach-Archiv (DHD 2016 – Komponisten-Datenbanken/-Portale: Austauschformate und Vernetzungspotentiale (Pre-conference Workshop)) in Leipzig (08.03)

Automatic Detection of Illuminated Charters (Talk)
at Uni Graz (Illuminated Charters – from the Margins of two Disciplines to the Core of Digital Humanities) in Wien (12.09)
From Script to Type Identification (Talk)
at Berlin-Brandenburgische Akademie der Wissenschaften (Workshop) in Berlin (06.09)

Weichselbaumer, Nikolaus; Christlein, Vincent
Automatische Typenbestimmung in historischen Drucken (Talk)
at Universität Leipzig (Digital Humanities im deutschsprachigen Raum (DHd 2016)) in Leipzig (09.03)

2015
Computergestützte Verfahren zur Analyse von hochmittelalterlichen Papsturkunden (Talk)
at FAU (Papstgeschichte des hohen Mittelalters: digitale und hilfswissenschaftliche Zugangsweisen zu einer Kulturgeschichte Europas) in Erlangen (20.02)
at Universität Leipzig (Doctoral Seminar) in Leipzig (05.11)
Technische Hilfsmittel zur Analyse von Papsturkunden des hohen Mittelalters (Talk)
at FAU (Interdisziplinäres Zentrum für Editionswissenschaften (IZED)) in Erlangen (26.01)
(13th International Conference on Document Analysis and Recognition (ICDAR)) in Nancy, France (24.08)
2014
at Universität Trier (Tagung – Möglichkeiten der Automatischen Manuskriptanalyse) in Trier (25.02)
at IEEE (2014 IEEE Winter Conference on Applications of Computer Vision (WACV)) in Steamboat Springs, CO (26.03)

 

2013

Christlein, Vincent
(A Letter Driven) Writer Identification in Medieval Papal Charters (Talk)
at University of Leeds (International Medieval Congress 2013) in Leeds (02.07)

Script and Signs. A Computer-based Analysis of High Medieval Papal Charters. A Key to Europe’s Cultural History (Talk)
at FAU (Workshop on Automatic Pattern Recognition and Historical Document Analysis) in Erlangen (14.06)
Writer Identification in Medieval Papal Charters (Talk)

at King’s College London (DigiPal Symposium III) in London (16.09)

2012

Christlein, Vincent
Handschrift und EDV – Möglichkeiten der Mustererkennung für paläographische Studien (Talk)
at FAU Erlangen-Nürnberg (Zugangsweisen zur päpstlichen Schriftlichkeit) in Lehrstuhl für Mittelalterliche Geschichte und Historische Hilfswissenschaften (14.07)

2010

Christlein, Vincent
A Study on Features for the Detection of Copy-Move Forgeries (Talk)
at Gesellschaft für Informatik (Sicherheit 2010 – Sicherheit, Schutz und Zuverlässigkeit) in Berlin (06.10)

On Rotation Invariance in Copy-Move Forgery Detection (Talk)

at IEEE (2nd IEEE Workshop on Information Forensics and Security (WIFS)) in Seattle, USA (15.12)

Vorlesung mit Übung (V/UE)

Kolloquium (KO)

Vorlesung (VORL)

Praktikum (PR)

  • Projekt Computer Vision

    This course will be held online until the coronavirus pandemic is contained to such an extent that the Bavarian state government can allow face-to-face teaching again

    Basic knowledge of image processing is desirable. In the first session there will be a short recap on image representation and basic image filtering techniques. However, having visited lectures such as Introduction to Pattern Recognition (IntroPR) or Diagnostic Medical Image Processing (DMIP) might prove beneficial.

    Please contact us if you have any questions. You can register via Studon (https://www.studon.fau.de/crs4397541.html) for the Computer Vision Project. During the semester lecture and exercise alternate on a weekly basis. Exercises are supervised and take place in one of the CIP pools. All exercises must be completed.

    You can get either 5 or 10 ECTS credits for this project. The following options are available:

    5 ECTS (counts as: Hochschulpraktikum)

    This option requires:

    • lectures (strongly recommended as they introduce the background required for the exercises)
    • exercises (in groups of 2 people) need to be finished on time
    • individual presentation about a state-of-the-art research paper at the end of the semester (graded if needed)

    10 ECTS (counts as Hochschulpraktikum (5 ECTS) + Forschungspraktikum (5 ECTS), or Master Project Computer Science (10 ECTS))

    • lectures (strongly recommended as they introduce the background required for the exercises)
    • exercises (in groups of 2 people) need to be finished on time
    • individual coding/research project under supervision of a LME PhD student at the end of regular schedule (graded if needed)

    Important: You cannot use the lecture/exercise part as a 5 ECTS research project (Forschungspraktikum). Please contact one of the PhD students at the lab if you need a research project.

    • Expected participants: 14; ECTS-Studium (ECTS-Credits: 10)
    • Termin:
      • Mo 12:00-14:00, Room Übung 3 / 01.252-128 (exclude vac) ICS
      • Mo 12:00-14:00, Room 00.156-113 CIP (exclude vac) ICS

Seminar (SEM)

  • Seminar Advanced Deep Learning

    • 2 SWS; Expected participants: 10; ECTS-Studium (ECTS-Credits: 5), Unterrichtssprache Englisch
    • Termin:
      • Di 12:00-14:00, Room Seminarraum ZMPT (exclude vac) ICS

Open

Please see https://lme.tf.fau.de/teaching/thesis/

Running

Title Type Start/end date Student
Writer Verification/Identification using SuperPoint and SuperGlue BA thesis May 16, 2022 - October 17, 2022 Alexander Klingebiel
Human interpretable Writer Retrieval and Verification MA thesis March 1, 2022 - August 31, 2022 Tim Raven
PowerPoint Presentation describer. Machine learning methods to automatically generate business captions from graphics MA thesis May 1, 2022 - November 1, 2022 Jovial Silatsa Tchatchum
Automated detection and defect recognition of photovoltaic modules in photoluminescence videos MA thesis March 1, 2022 - September 1, 2022 Susmitha Rachamreddy
Fruit Terminator – Annotation of Lung Fluid Cells via Gamification BA thesis January 10, 2022 - June 10, 2022 Marco Martin Härtl
Letter Inpainting and Detection of Mathematical Diagrams in Multi-Lingual Manuscripts using a Deep Neural Network approach BA thesis January 1, 2022 - June 1, 2022 Sebastian Baum
Determining the Influence of Papyrus Characteristics on Fragments Retrieval with Deep Metric Learning MA thesis December 1, 2021 - June 1, 2022 Timo Bohnstedt
Multi-stage Patch based U-Net for Text Line Segmentation of Historical Documents MA thesis December 1, 2021 - June 1, 2022 Shrihari Muttagi
Deep Learning-based Bleed-through Removal in Historical Documents MA thesis November 1, 2021 - May 1, 2022 Vojtech Pesek
Terahertz Image Reconstruction for Historical Document Analysis MA thesis November 5, 2021 - May 4, 2022 Balaka Dutta

Finished

Bachelor’s Theses: 6
Master’s Theses: 20

Friedrich-Alexander-Universität Erlangen-Nürnberg