ICDAR 2017 Competition on Historical Document Writer Identification (Historical-WI)


Scoreboard

Position Name Method Info Submitter Affiliation Submitted before deadline Score
1 Tebessa_oBIFColfilSE In this method, the different configurations of oBIFs columns histograms extracted from smoothed binary historical document samples with low-pass filters are concatenated for generating feature vector and the maximum product of Spearman and Euclidean distance used for classifying the historical docs Abdeljalil GATTAL, Chawki Djeddi Larbi Tebessi University, Tebessa, Algeria, Department of Mathematics and Computer Science, Larbi Tebessi University, Tebessa 3
2 CoHinge This method uses coHinge feature for writer identification, which is described on the paper "Beyond OCR: Multi-faceted understanding of handwritten document characteristics." pattern recognition Sheng He, neng he Artificial Intelligence and Image Analysis, University of Groningen 5
3 Tebessa_oBIFColfiltS In this method, the different configurations of oBIFs columns histograms extracted from smoothed binary historical document samples with low-pass filters are concatenated for generating feature vector and the maximum product of Spearman and Euclidean distance used for classifying the historical docs Abdeljalil GATTAL, Chawki Djeddi Larbi Tebessi University, Tebessa, Algeria, Department of Mathematics and Computer Science, Larbi Tebessi University, Tebessa 7
4 NLNBNN_FAST5 A classifier for offline writer identification based on the Local Na¨ıve Bayes Nearest-Neighbour (Local NBNN) classification. It takes into consideration the particularity of handwriting patterns by adding a constraint to prevent the matching of irrelevant keypoints. Hussein Mohammed Centre for the study of manuscript cultures (CSMC) - Hamburg University 7
5 Tebessa_oBIFColfilt1 In this method, the different configurations of oBIFs columns histograms extracted from smoothed binary historical document samples with low-pass filters are concatenated for generating feature vector and City block distance measures used for classifying historical document. Abdeljalil GATTAL, Chawki Djeddi Larbi Tebessi University, Tebessa, Algeria, Department of Mathematics and Computer Science, Larbi Tebessi University, Tebessa 9
6 Tebessa_oBIFColumns1 In this method, The different configurations of oriented Basic Image Features (oBIFs) columns histograms extracted from binarized historical document samples are concatenated for generating for generating feature vector and City block distance measures used for classifying historical document. Abdeljalil GATTAL, Chawki Djeddi Larbi Tebessi University, Tebessa, Algeria, Department of Mathematics and Computer Science, Larbi Tebessi University, Tebessa 11
7 Tebessa_HDWI_oBIF1 In this method, oBIFs histograms and oBIFs columns histograms extracted from binarized historical document samples are concatenated for generating feature vector and City block distance measures (also referred to as Manhattan distance measures) used for classifying historical document. Abdeljalil GATTAL, Chawki Djeddi Larbi Tebessi University, Tebessa, Algeria, Department of Mathematics and Computer Science, Larbi Tebessi University, Tebessa 14
8 Triplet_Network_WI_3 Revised submission format (fixed filenames). Vinaychandran Pondenkandath University of Fribourg 17
9 Triplet_Network_WI_4 Improved data augmentation (multi-crop and multi-scale) Vinaychandran Pondenkandath University of Fribourg 17
10 DEB113W_6 Debuggin 113 Apostolis Barbagiannis NCSR Demokritos / IIT / CIL 20
11 Triplet_Network_WI Our method uses a deep convolutional neural network (CNN), trained using the triplet margin loss metric to transform a given input into a space where inputs belonging to the same class (writer) are close to each other. We use triplets which consist of the anchor, positive and negative samples. Vinaychandran Pondenkandath University of Fribourg 22
12 Tebessa_HDWI_oBIF In this method, oBIFs histograms and oBIFs columns histograms extracted from binarized historical document samples are concatenated for generating feature vector and City block distance measures (also referred to as Manhattan distance measures) used for classifying historical document. Abdeljalil GATTAL, Chawki Djeddi Larbi Tebessi University, Tebessa, Algeria, Department of Mathematics and Computer Science, Larbi Tebessi University, Tebessa 24
13 Tebessa_oBIFColumns In this method, The different configurations of oriented Basic Image Features (oBIFs) columns histograms extracted from binarized historical document samples are concatenated for generating for generating feature vector and City block distance measures used for classifying historical document. Abdeljalil GATTAL, Chawki Djeddi Larbi Tebessi University, Tebessa, Algeria, Department of Mathematics and Computer Science, Larbi Tebessi University, Tebessa 26
14 Triplet_Network_WI_2 Revised submission format. Vinaychandran Pondenkandath University of Fribourg 28

News

Website online Dataset released change of the submission date

Important Dates

Release of the Dataset: 17.03.2017

Submission open 23.06.2017

Submission Deadline: 05.07.2017
12.07.2017





Organizers







Stefan Fiel

[TU Wien, Computer Vision Lab] 

Stefan Fiel works as research assistant at the Computer Vision Lab, TU Wien.

Basilis Gatos

[NCSR Demokritos / IIT / CIL] 

Researcher at the Institute of Informatics and Telecommunications of the National Center for Scientific Research "Demokritos", Athens, Greece.

Georgios Louloudis

[NCSR Demokritos / IIT / CIL] 

Research Associate at the Institute of Informatics and Telecommunications of the National Center for Scientific Research "Demokritos", Athens, Greece.

Nikolaos Stamatopoulos

[NCSR Demokritos / IIT / CIL] 

Research Associate at the Institute of Informatics and Telecommunications of the National Center for Scientific Research "Demokritos", Athens, Greece.

Markus Diem

[TU Wien, Computer Vision Lab] 

Markus Diem is a senior scientist at the Computer Vision Lab, TU Wien, Austria. His research interests are Cultural Heritage Applications and Document Analysis.

Florian Kleber

[TU Wien, Computer Vision Lab] 

Florian Kleber is currently a senior scientist at the Computer Vision Lab, Institute for Computer Aided Automation, TU Wien, Austria. His research interests are Cultural Heritage Applications and Document Analysis Applications.

Vincent Christlein

[FAU Erlangen-Nürnberg, Pattern Recognition Lab] 

Vincent Christlein is a PhD student at the Pattern Recognition Lab, FAU Erlangen-Nuremberg.