Conference
LADaS: A Diachronic Multi-Type Dataset for Document Layout Analysis from the 17th Century to the Present
| Τίτλος: | LADaS: A Diachronic Multi-Type Dataset for Document Layout Analysis from the 17th Century to the Present |
|---|---|
| Συγγραφείς: | Janes, Juliette, Bénière, Sarah, Kiessling, Benjamin, Ing, Lucence, Astier, Éric, Gabay, Simon, Sagot, Benoît, Clérice, Thibault |
| Συνεισφορές: | Automatic Language Modelling and ANAlysis & Computational Humanities (ALMAnaCH), Centre Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Persée, École normale supérieure de Lyon (ENS de Lyon), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS), Université de Genève = University of Geneva (UNIGE), DEFI COLaF, European Project: 101132163,HORIZON-INFRA-2023-SERV-01,HORIZON-INFRA-2023-SERV-01,ATRIUM(2024) |
| Πηγή: | 8th International Workshop on Historical Document Imaging and Processing (HIP’26), The 20th International Conference on Document Analysis and Recognition (ICDAR) https://hal.science/hal-05638101 8th International Workshop on Historical Document Imaging and Processing (HIP’26), The 20th International Conference on Document Analysis and Recognition (ICDAR), Sep 2026, Vienne, Austria |
| Στοιχεία εκδότη: | CCSD |
| Έτος έκδοσης: | 2026 |
| Συλλογή: | Université de Lyon: HAL |
| Θεματικοί όροι: | Historical Documents, Cross-domain Transfer, Diachronic Dataset, Object Detection, Document Layout Analysis, Annotation Guidelines, Text Encoding Initiative, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], [SHS.HIST]Humanities and Social Sciences/History, [SHS.LITT]Humanities and Social Sciences/Literature, [SHS.MUSEO]Humanities and Social Sciences/Cultural heritage and museology |
| Θέμα γεωγραφικό: | Vienne, Austria |
| Περιγραφή: | International audience ; We introduce an open-access and openly-licensed diachronic dataset for Document Layout Analysis (DLA) targeting heterogeneous historical and contemporary materials 4 . The dataset comprises 8,287 manually annotated pages spanning 1600-2024, covering both digitised and born-digital documents across diverse genres including magazines, academic papers, monographs, plays, and administrative records. Annotations follow a two-level vocabulary derived from SegmOnto, extended to paragraph-level semantic regions consistent with the Text Encoding Initiative (TEI), and provide 45 fine-grained classes over 115,007 instances. The data is organised in modular subsets, enabling domain-specific configurations. We establish baselines using two families of object detectors (YOLOv11 and D-FINE) and study three regimes: in-domain training, label-set simplification, and large-scale pretraining (Objects365). We additionally conduct a cross-dataset transfer study against DocLayNet and M 6 Doc using a unified label mapping and a per-decade temporal evaluation. Results show that models trained on existing modern-centric DLA datasets degrade sharply on pre-20th century material (mean F1 dropping by up to 0.25), while in-domain training on our dataset preserves performance across centuries. This highlights the need for diachronic benchmarks and demonstrates the value of combining CV-based detection with DH-oriented annotation standards. |
| Τύπος εγγράφου: | conference object |
| Γλώσσα: | English |
| Relation: | info:eu-repo/grantAgreement//101132163/EU/Advancing FronTier Research In the Arts and hUManities/ATRIUM |
| Διαθεσιμότητα: | https://hal.science/hal-05638101 https://hal.science/hal-05638101v1/document https://hal.science/hal-05638101v1/file/HIP-4.pdf |
| Rights: | https://creativecommons.org/licenses/by/4.0/ ; info:eu-repo/semantics/OpenAccess |
| Αριθμός Καταχώρησης: | edsbas.4DD5571B |
| Βάση Δεδομένων: | BASE |
| Η περιγραφή δεν είναι διαθέσιμη |