Technical Program

Friday, November 10, 2017

09:00 Opening, Welcome by Andreas Fischer, Chair HIP’17
Session 1: Character Datasets and Text Retrieval
Chair: Liangrui Peng
09:15 A New Khmer Palm Leaf Manuscript Dataset for Document Analysis and Recognition – SleukRith Set
Dona Valy, Michel Verleysen, Sophea Chhun and Jean-Christophe Burie
09:35 Isolated Character Forms from Dated Syriac Manuscripts
Nicholas Howe, Minyue Dai and Michael Penn
09:55 Towards Letter Shape Prior and Paleographic Tables Estimation in Hebrew First Temple Period Ostraca
Arie Shaus and Eli Turkel
10:15 Text Retrieval for Japanese Historical Documents by Image Generation
Chisato Sugawara, Tomo Miyazaki, Yoshihiro Sugaya and Shinichiro Omachi
10:35 A Handwritten French Dataset for Word Spotting – CFRAMUZ
Nikolaos Arvanitopoulos, Gaspard Chevassus, Daniele Maggetti and Sabine Süsstrunk
10:55 Coffee Break
Session 2: Character Recognition
Chair: Nicholas Howe
11:15 Attempts to recognize anomalously deformed Kana in Japanese historical documents
Hung Tuan Nguyen, Nam Tuan Ly, Kha Cong Nguyen, Cuong Tuan Nguyen and Masaki Nakagawa
11:35 Tens of Thousands of Nom Character Recognition by Deep Convolution Neural Networks
Kha Cong Nguyen, Cuong Tuan Nguyen and Masaki Nakagawa
11:55 Convolutional Neural Network with Attention Mechanism for Historical Chinese Character Recognition
Haoyu Qin and Liangrui Peng
12:15 Training LSTM-RNN with Imperfect Transcription – Limitations and Outcomes
Martin Jenckel, Syed Saqib Bukhari and Andreas Dengel
12:35 Methods of data augmentation for palimpsest character recognition with Deep Neural Network
Anna Starynska, Roger L. Jr. Easton and David Messinger
12:55 Lunch Break
Session 3: Layout Analysis and Complete Workflows
Chair: Bertrand Coüasnon
14:30 PageNet: Page Boundary Extraction in Historical Handwritten Documents
Christopher Tensmeyer, Brian Davis, Curtis Wigington, Iain Lee and Bill Barrett
14:50 Mass Digitization of Archival Documents using Mobile Phones
Florian Kleber, Markus Diem, Fabian Hollaus and Stefan Fiel
15:10 Robust Heartbeat-based Line Segmentation Methods for Regular Texts and Paratextual Elements
Mathias Seuret, Daniel Stökl Ben Ezra and Marcus Liwicki
15:30 Deep Convolutional Neural Networks for Image Resolution Detection
Felix Trier, Muhammad Zeshan Afzal, Markus Ebbecke and Marcus Liwicki
15:50 Creating a Complete Workflow for Digitising Historical Census Documents: Considerations and Evaluation
Christian Clausner, Justin Hayes, Apostolos Antonacopoulos and Stefan Pletschacher
16:10 Coffee Break
Session 4: Binarization and Page Segmentation
Chair: Apostolos Antonacopoulos
16:30 Automatic Document Image Binarization using Bayesian Optimization
Ekta Vats, Anders Hast and Prashant Singh
16:50 Historical Document Image Segmentation with LDA-Initialized Deep Neural Networks
Michele Alberti, Mathias Seuret, Vinaychandran Pondenkandath, Rolf Ingold and Marcus Liwicki
17:10 Document Image Page Segmentation and Character Recognition as Semantic Segmentation
Seth Stewart and Bill Barrett
17:30 HBA 1.0: A Pixel-based Annotated Dataset for Historical Book Analysis
Maroua Mehri, Pierre Héroux, Rémy Mullot, Jean-Philippe Moreux, Bertrand Coüasnon and Bill Barrett
17:50 ICDAR 2017 HBA Competition
Maroua Mehri, Pierre Héroux, Rémy Mullot, Jean-Philippe Moreux, Bertrand Coüasnon and Bill Barrett
18:00 IAPR Best Paper Award
18:05 Closing by Bill Barrett, General Chair HIP’17