Result Details

Archival Faces: Detection of Faces in Digitized Historical Documents

VAŠKO, M.; HEROUT, A.; HRADIŠ, M. Archival Faces: Detection of Faces in Digitized Historical Documents. In Document Analysis and Recognition – ICDAR 2025 Workshops: Wuhan, China, September 20–21, 2025, Proceedings, Part II. Cham: Springer Nature Switzerland, 2026. p. 17-34. ISBN: 978-3-032-09370-7.
Type
conference paper
Language
English
Authors
Vaško Marek, Ing., DCGM (FIT)
Herout Adam, prof. Ing., Ph.D., DCGM (FIT)
Hradiš Michal, Ing., Ph.D., UAMT (FEEC), DCGM (FIT)
Abstract

When digitizing historical archives, it is necessary to search for the faces of celebrities and ordinary people, especially in newspapers, link them to the surrounding text, and make them searchable. Existing face detectors on datasets of scanned historical documents fail remarkably -- current detection tools only achieve around 24% mAP at 50:90% IoU. This work compensates for this failure by introducing a new manually annotated domain-specific dataset in the style of the popular Wider Face dataset, containing 2.2k new images from digitized historical newspapers from the 19th to 20th century, with 11k new bounding-box annotations and associated facial landmarks. This dataset allows existing detectors to be retrained to bring their results closer to the standard in the field of face detection in the wild. We report several experimental results comparing different families of fine-tuned detectors against publicly available pre-trained face detectors and ablation studies of multiple detector sizes with comprehensive detection and landmark prediction performance results.

Keywords

historoical documents, face detection, object detection, biometry

URL
Published
2026
Pages
17–34
Proceedings
Document Analysis and Recognition – ICDAR 2025 Workshops: Wuhan, China, September 20–21, 2025, Proceedings, Part II
Conference
International Conference on Document Analysis and Recognition
ISBN
978-3-032-09370-7
Publisher
Springer Nature Switzerland
Place
Cham
DOI
EID Scopus
BibTeX
@inproceedings{BUT197677,
  author="Marek {Vaško} and Adam {Herout} and Michal {Hradiš}",
  title="Archival Faces: Detection of Faces in Digitized Historical Documents",
  booktitle="Document Analysis and Recognition – ICDAR 2025 Workshops: Wuhan, China, September 20–21, 2025, Proceedings, Part II",
  year="2026",
  pages="17--34",
  publisher="Springer Nature Switzerland",
  address="Cham",
  doi="10.1007/978-3-032-09371-4\{_}2",
  isbn="978-3-032-09370-7",
  url="https://link.springer.com/chapter/10.1007/978-3-032-09371-4_2"
}
Projects
Reanimated book - digitized library treasures for the creative industries, MK, NAKI III – program na podporu aplikovaného výzkumu v oblasti národní a kulturní identity na léta 2023 až 2030, DH23P03OVV033, start: 2023-03-01, end: 2027-12-31, running
Departments
Back to top