The 16th International Conference on
Frontiers in Handwriting Recognition
August 5 - 8, 2018 ● Niagara Falls, USA
-Keynotes -Talks -Posters -Tutorials -Competitions
-Instructions for Presenters
- Keynote Talks -
Deciphering Historical Manuscripts
Speaker: Kevin Knight
Abstract. European libraries and archives are filled with enciphered manuscripts from the early modern period. These include military and diplomatic correspondence, records of secret societies, private letters, and so on. Even though they are enciphered with older cryptographic methods, their contents are still unavailable to working historians. In this talk, we will describe such manuscripts and show how computer algorithms can help decipher them, both from digital transcriptions and from raw pixel images.
Short Biography. Kevin Knight is Dean's Professor of Computer Science at the University of Southern California (USC) and Chief Scientist for Natural Language Processing at Didi Chuxing. He is best known for his contributions to statistical machine translation, natural language generation, automata theory, and decipherment of historical manuscripts. Prof. Knight is a Fellow of ACL, AAAI, and ISI.
Title: TBAAbstract. TBA.
Speaker: Gregory R. Crane
Short Biography. Gregory R. Crane is the Alexander von Humboldt Professor of Digital Humanities at the University of Leipzig and Winnick Family Chair of Technology and Entrepeneurship at Tufts University. Greg owes his reputation as a pioneer of digital humanities to his development of the Perseus Digital Library, a comprehensive, freely accessible online library for antique source material. He has received, among other awards, the Google Digital Humanities Award 2010 for his work in the field.
- Oral Sessions -
Monday, August 6th
Oral Session 1: Document Segmentation and Understanding
23 A CNN-based Approach to Detecting Text from Images of Whiteboards and Handwritten Notes Wei Jia, Lei Sun, Zhuoyao Zhong and Qiang Huo 34 dhSegment: A generic deep-learning approach for document segmentation Sofia Ares Oliveira, Benoit Seguin and Frederic Kaplan 80 Character and Text Recognition of Khmer Historical Palm Leaf Manuscripts Dona Valy, Michel Verleysen, Sophea Chhun and Jean-Christophe Burie 123 Automated Detection of Handwritten Whiteboard Content on Lecture Videos for Summarization Bhargava Urala Kota, Kenny Davila, Alexander Stone, Srirangaraj Setlur and Venu Govindaraju 142 Recognizing Challenging Handwritten Annotations with Fully Convolutional Networks Andreas Kölsch, Ashutosh Mishra, Saurabh Varshneya and Marcus Liwicki
Oral Session 2: Word/Object Spotting
45 Towards Spotting and Recognition of Handwritten Documents in Indic Scripts Kartik Dutta, Praveen Krishnan, Minesh Mathew and C V Jawahar 68 A Probabilistic Retrieval Model for Word Spotting based on Direct Attribute Prediction Eugen Rusakov, Leonard Rothacker, Hyunho Mo and Gernot Fink 115 Probabilistic Indexing and Search for Information Extraction on Handwritten German Parish Records Eva Lang, Joan Puigcerver, Alejandro Toselli and Enrique Vidal 124 Visual Search Engine for Handwritten and Typeset Math in Lecture Videos and LaTeX Notes Kenny Davila and Richard Zanibbi
Oral Session 3: Data Augmentation
60 Distilling GRU with Data Augmentation for Unconstrained Handwritten Text Recognition Manfei Liu, Zecheng Xie, Yaoxiong Huang, Lianwen Jin and Weiying Zhou 75 Memory-Augmented Attention Model for Scene Text Recognition Cong Wang, Fei Yin and Cheng-Lin Liu 81 Generating Handwritten Character Clones from an Incomplete Seed Character Set using Collaborative Filtering Kazuaki Nakamura, Eiji Miyazaki, Naoko Nitta and Noboru Babaguchi 106 Training an End-to-end Model for Offline Handwritten Japanese Text Recognition by Generated Synthetic Patterns Nam Tuan Ly, Cuong Tuan Nguyen and Masaki Nakagawa
Tuesday, August 7th
Oral Session 4: Signature/Writer Verification
24 Robotic Arm Motion for Verifying Signatures Moises Diaz, Miguel Ferrer and Jose J. Quintana 25 Offline Signature Verification via Structural Methods: Graph Edit Distance and Inkball Models Paul Maergner, Nicholas R. Howe, Kaspar Riesen, Rolf Ingold and Andreas Fischer 58 Histogram-based matching of GMM encoded features for online signature verification Suresh Sundaram and Abhishek Sharma 65 Learning Discriminative Feature Hierarchies for Off-line Signature Verification Songxuan Lai and Lianwen Jin 77 Writer verification using CNN feature extraction Jun Chu, Mohammad Abuzar Shaikh, Mihir Chauhan, Lu Meng and Sargur Srihari 79 Hybrid Feature Learning for Handwriting Comparison Mohammad Abuzar Shaikh, Mihir Chauhan, Jun Chu and Sargur Srihari
Oral Session 5: Recurrent Neural Networks for Character Recognition
76 DenseRAN for Offline Handwritten Chinese Character Recognition Wenchao Wang, Jianshu Zhang, Jun Du, Zi-Rui Wang and Yixing Zhu 99 Accelerating and Compressing LSTM based Model for Online Handwritten Chinese Character Recognition Yafeng Yang, Kaihuan Liang, Xuefeng Xiao, Zecheng Xie, Lianwen Jin, Jun Sun and Weiying Zhou 118 From HMMs to RNNs: Computer-assited Transcription of a Handwritten Notarial Records Collection Lorenzo Quirós, Vicente Bosch, Lluis Serrano, Alejandro Toselli and Enrique Vidal 120 Transcription Free LSTM OCR-Model Evaluation Martin Jenckel, Syed Saqib Bukhari and Andreas Dengel 122 Boosting the deep multidimensional long-short-term memory network for handwritten recognition systems Dayvid Castro, Byron L. D. Bezerra and Mêuser Valença
Wednesday, August 8th
Oral Session 6: Transfer Learning
8 Language Model Supervision for Handwriting Recognition Model Adaptation Christopher Tensmeyer, Curtis Wigington, Brian Davis, Seth Stewart, Tony Martinez and William Barrett 22 Building Compact CNN-DBLSTM based Character Models for Handwriting Recognition and OCR by Teacher-Student Learning Haisong Ding, Kai Chen, Wenping Hu, Meng Cai and Qiang Huo 54 Parsimonious HMMs for Offline Handwritten Chinese Text Recognition Wenchao Wang, Jun Du and Zi-Rui Wang 69 Deep Transfer Mapping for Unsupervised Writer Adaptation Hong-Ming Yang, Xu-Yao Zhang, Fei Yin, Jun Sun and Cheng-Lin Liu
Oral Session 7: CNNs for Handwriting Recognition
1 Improving CNN-RNN Hybrid Networks for Handwriting Recognition Kartik Dutta, Praveen Krishnan, Minesh Mathew and C V Jawahar 42 Fully Convolutional Networks for Handwriting Recognition Felipe Petroski Such, Dheeraj Kumar Peri, Frank Brockler, Paul Hutkowski and Raymond Ptucha 70 Introducing Local Distance-based Features to Temporal Convolutional Neural Networks Brian Kenji Iwana, Minoru Mori, Akisato Kimura and Seiichi Uchida 104 Discovering Class-wise Trends of Max-pooling in Subspace Yuchen Zheng, Brian Kenji Iwana and Seiichi Uchida
- Poster Sessions -
Poster Session 1 - Monday, August 6th
18 Recognition of Chinese Text in Historical Documents with Page-Level Annotations Hailin Yang and Lianwen Jin 26 Rotation-free Handwritten Chinese Character Recognition Using Two-stage Convolutional Neural Network Zhe Li, Lianwen Jin and Songxuan Lai 28 Learning Features for Writer Retrieval and Identification using Triplet CNNs Manuel Keglevic and Stefan Fiel 30 Matching Table Structures of Historical Register Books using Association Graphs Florian Kleber, Markus Diem, Herve Dejean, Jean-Luc Meunier and Eva Lang 31 2DLDA-based Compound Distance for Similar Online Handwritten Tibetan Transliteration of the Sanskrit Character Recognition Zhengqi Cai and Weilan Wang 35 Improving Word Spotting System Performance Using Ensemble Classifier Combination Methods Muna Khayyat and Ching Suen 46 Localizing and Recognizing Text in Lecture Videos Kartik Dutta, Minesh Mathew, Praveen Krishnan and C V Jawahar 51 A Hybrid Method for Text Line Extraction in Handwritten Document Images Ehsan Kiumarsi and Alireza Alaei 52 Crowdsourcing Online Handwriting Acquisition to Develop and Deploy a Unicode Character Classifier Francisco Alvaro and Daniel Martín-Albo 55 Word Beam Search: A Connectionist Temporal Classification Decoding Algorithm Harald Scheidl, Stefan Fiel and Robert Sablatnig 73 A deep Convolutional Encoder-Decoder Network for Page Segmentation of Historical Handwritten Documents into Text Zones Panagiotis Kaddas and Basilis Gatos 74 Neural Text Line Segmentation of Multilingual Print and Handwriting with Recognition-Based Evaluation Patrick Schone, Christian Hargraves, Jon Morrey, Rachael Day and Mindy Jacox 86 Real-time analysis of hand-drawn sketches with extended bi-dimensional grammar Omar Krichen, Nathalie Girard, Eric Anquetil, Simon Corbillé and Mickaël Renault 87 Automated Grading of Handwritten Essays Annapurna Sharma and Dinesh Babu Jayagopi 97 Multi-perspective Multi-modal Trajectory Descriptions for Handwritten Strokes Mohammad Tanvir Parvez and Sardar Anisul Haque 101 Active learning in Handwritten Text Recognition using the derivational entropy Verónica Romero and Joan Andreu Sánchez 102 Training Schemes for the Transliteration of the Balinese Script into the Latin Script on Palm Leaf Manuscript Images Made Windu Antara Kesiman, Jean-Christophe Burie, Jean-Marc Ogier and Philippe Grange 107 Watermarking for Security Issue of Handwritten Documents with Fully Convolutional Networks Vinh Loc Cu, Jean-Christophe Burie and Jean-Marc Ogier 108 Separating Optical and Language Models through Encoder-Decoder Strategy for Transferable Handwriting Recognition Adeline Granet, Emmanuel Morin, Harold Mouchère, Solen Quiniou and Christian Viard-Gaudin 110 Compact Deep Decriptors for Keyword Spotting George Retsinas, Giorgos Sfikas, Georgios Louloudis, Nikolaos Stamatopoulos and Basilis Gatos 116 DNN-HMM based Large Vocabulary Online Handwritten Assamese Word Recognition System Subhasis Mandal, Himakshi Choudhury, S. R. Mahadeva Prasanna and Suresh Sundaram 119 Constrained and Parametric Dynamic Programming for Word Image retrieval Tanmoy Mondal 135 Identifying Cross Depicted Motifs in Historical Handwritten Documents Vinay Pondenkandath, Michele Alberti, Nicole Eichenberger, Rolf Ingold and Marcus Liwicki 138 An Investigative Analysis of Different LSTM Libraries for Supervised and Unsupervised Architectures of OCR Training Syed Saqib Bukhari, Sumam Francis and Andreas Dengel
Poster Session 2 - Tuesday, August 7th
3 A Study on Idiosyncratic Handwriting with Impact on Writer Identification Chandranath Adak, Bidyut B. Chaudhuri and Michael Blumenstein 4 Extracting Stroke Errors from Digital Ink Characters by Beginning Learners of Chinese as A Foreign Language Based on Acurate Stroke Matching Hao Bai and Xi-Wen Zhang 7 An effective Binarization method for disturbed camera-captured document images Jinyuan Zhao, Cunzhao Shi, Fuxi Jia, Yanna Wang and Baihua Xiao 14 Zero-Shot Learning Based Approach For Medieval Word Recognition Using Deep-Learned Features Sukalpa Chanda, Jochem Baas, Daniël Haitink, Sebastien Hamel, Dominique Stutzmann and Lambert Schomaker 16 Tracking the Ballistic Trajectory in Complex and Long Handwritten Signatures Gioele Crispo, Moises Diaz, Angelo Marcelli and Miguel Angel Ferrer 29 Text Line Extraction Based on Distance Map Features and Dynamic Programming Vicente Bosch Campos, Verónica Romero, Alejandro Toselli and Enrique Vidal 32 Feature Descriptors for Spotting 3D Characters on Triangular Meshes Bartosz Bogacz and Hubert Mara 39 Writer Identification on Historical Documents Using Oriented Basic Image Features Abdeljalil Gattal, Chawki Djeddi, Imran Siddiqi and Somaya Al-Maadeed 41 Text Line Segmentation for Challenging Handwritten Document Images using Fully Convolutional Network Berat Barakat, Ahmad Droby, Majeed Kassis and Jihad El-Sana 43 Introducing the XXX Bangla Handwriting Dataset and an Efficient Offline Recognizer of Isolated Bangla Characters Nishatul Majid and Elisa Barney Smith 49 New RGB based Fusion for Forged IMEI Number Detection in Mobile Images Palaiahnakote Shivakumara, V. Basavaraja, Harsha S. Gowda, D. S. Guru, Umapada Pal and Tong Lu 50 Adaptive Multi-Gradient Kernels for Gender Identification B. J Navya, Shivakumara Palaiahnakote, G.C Shwetha, Sangheeta Roy, D. S. Guru, Umapada Pal and Tong Lu 59 A Hierarchical Codebook Descriptor Approach for Online Writer Identification Suresh Sundaram, Vivek Venugopal and Surbhi Pillai 63 A CNN Based Framework for Multi-Lingual Unistroke Numeral Recognition in Air-Writing Prasun Roy, Subhankar Ghosh and Umapada Pal 82 Exploring Sparse Representation for Improved Online Handwriting Recognition Subhasis Mandal, Sayed Shahnawazuddin, Rohit Sinha, S R Mahadeva Prasanna and Suresh Sundaram 83 Signature and Logo Detection using Deep Convolutional Neural Networks Nabin Sharma, Ranju Mandal, Rabi Sharma, Umapada Pal and Michael Blumenstein 94 DeepDIVA: A Highly-Functional Python Framework for Reproducible Experiments Michele Alberti, Vinay Pondenkandath, Marcel Würsch, Rolf Ingold and Marcus Liwicki 103 Boosting Handwriting Text Recognition in Small Databases with Transfer Learning José Carlos Aradillas Jaramillo, Juan José Murillo Fuentes and Pablo Martínez Olmos 105 Online Japanese Handwriting Recognizers using Recurrent Neural Networks Hung Tuan Nguyen, Cuong Tuan Nguyen and Masaki Nakagawa 128 CNN based Segmentation of Text-lines from Handwritten/Printed Document Images Arpita Dutta, Arpan Garai and Samit Biswas 131 Offline Text-Independent Writer Identification Based on Writer-Independent Model Using Conditional AutoEncoder Mariko Hosoe, Tomoki Yamada, Kunihito Kato and Kazuhiko Yamamoto 147 Angle Minimization and Graph Analysis for text line segmentation in handwritten documents Insaf Setitra and Abdelkrim Meziane 151 Multimodal acquisition and analysis of children handwriting for the study of the efficiency of their handwriting movements: the @MaGma challenge Remi Celine, Jimmy Nagau, Jean Vaillant, Alin Dorville and Réjean Plamondon 154 Segmentation and Stitching Improves Handwriting Recognition on Datasets with Few Samples Seth Stewart, Lucas Pinto and William Barrett
Poster Session 3 - Wednesday, August 8thNote: Competition Posters TBA
13 Pixel-level Reconstruction and Classification for Noisy Handwritten Bangla Characters Manohar Karki, Supratik Mukhopadhyay, Saikat Basu, Robert Dibiano and Qun Liu 20 A Benchmark Dataset of Online Handwritten Gurmukhi Script Words and Numerals Harjeet Singh, R.K. Sharma, Rajesh Kumar, Karun Verma, Ravinder Kumar and Munish Kumar 33 RNN - Based Online Handwritten Word Recognition in Devanagari Script Rajib Ghosh, Pooja Kesari and Prabhat Kumar 48 Handwritten Arabic character recognition using features extracted by blind source separation Mohamed Ali Souibgui, Samir Belaid and Wady Naanaa 56 New COLD Feature based Handwriting Analysis for Enthnicity/Nationality Identification Sauradip Nag, Palaiahnakote Shivakumara, Yirui Wu, Umapada Pal and Tong Lu 64 Word-Hunter: A Gamesourcing Experience to Validate the Transcription of Historical Manuscripts Jialuo Chen, Pau Riba, Alicia Fornés, Joan Mas, Josep Llados and Joana Maria Pujadas 66 Writer Identification for Historical Manuscripts: Analysis and Optimisation of a Classifier as an Easy-to-Use Tool for Scholars from the Humanities Hussein Mohammed, Volker Märgner and H. Siegfried Stiehl 71 A Fully Convolutional Network for Signature Segmentation Victor Kléber Santos Leite Melo and Byron Bezerra 78 Hand-Written and Machine-Printed Text Classification in Architecture, Engineering & Construction Documents Supriya Das 84 A Study of Data Augmentation for Handwritten Character Recognition Using Deep Learning Taihei Hayashi, Keiji Gyohten, Hidehiro Ohki and Toshiya Takami 90 Probabilistic Music-Symbols Spotting in handwritten scores Jorge Calvo-Zaragoza, Alejandro Toselli and Enrique Vidal 92 Data Driven Feature Extraction for Gender Classification using Multi-script Handwritten Texts Momina Moetesum, Imran Siddiqi, Chawki Djeddi, Yaacoub Hannad and Somaya Al-Maadeed 111 MultiSpectral Image Binarization using GMMs Fabian Hollaus, Markus Diem and Robert Sablatnig 114 Space Balancing in Online Handwriting Recognition Postprocessing Using Deep Bidirectional LSTM Olha Zubarieva, Ivan Deriuga, Vadym Holosko and Tetiana Ignatova 141 A New Template Matching Similarity Based On Shape Context Descriptor Nabil Aouadi and Afef Kacem 143 Hybrid two-dimensional recognizer based on the NSHP-HMM model Hanene Boukerma, Christophe Choisy, Abdallah Benouareth, Nadir Farah and Mohamed Cheriet 156 Finding the answer: techniques for locating students' answers in handwritten problem solutions Amirali Darvishzadeh, Negin Entezari and Thomas Stahovich Important Notes
- Instructions for Presenters -
- If a paper is not presented (oral or poster) by one of its authors, then the paper will be withdrawn from the proceedings and from IEEE Xplore. There will be NO exceptions to this policy.
- Authors may not delegate presentation of their paper to anyone else.
- You must ensure that your presentation does not exceed 15 minutes. This will give 5 minutes for the session chair and audience to pose questions and have discussions. Please meet the session chair for your session at least 15 minutes before the session begins.
- If you are using your own laptop for the presentation, please ensure that it is compatible with the projector present in the room before the session. As a precautionary measure, please also bring a USB memory stick containing your presentation.
- If you will be using the laptop present in the conference hall, please bring a USB memory stick containing your presentation. The laptop will have Microsoft PowerPoint and Acrobat available in it.
- Wi-fi access to the internet will be available in the conference hall.
- Contact email@example.com if you have any queries not answered on this page.
- Each poster display board will be of size 8ft (width) x 4ft (height). and any poster that fits within that dimension is fine.
- Posters with landscape orientation preferred. Ideal dimensions: A0 (1189 mm wide x 841 mm high) or 4ft wide by 3 ft high
- All posters can be mounted from 8:00 am on the day of the presentation, and must be necessarily mounted before the poster session begins. Mounting boards and pins/adhesives will be available in the poster room.
- The posters in each session will be numbered. The list of numbers will be posted next to the board. Please use your numbered location.
- One of the authors must be present at their poster for the entire duration of their session.
- We are exploring the possibility of having the posters on display on all days of the conference. If this were to happen, poster authors will be notified via email prior to the start of the conference to mount their posters before the start of the first poster session on the first day.
August 5-8, 2018
Niagara Falls (USA)