ASAR2021 Competition on Online Arabic Word Database ADAB
ABSTRACT
Online Handwriting recognition still a challenging field of research. Several investigations are still necessary due to the cursiveness of some scripts. In this context, it handles the problem of Arabic online handwritten script which still under-exploited compared to Latin script.
In order to advance the research and to raise online handwriting recognition performance of online Arabic script, we organize the ASAR2021 Competition on Online Arabic Words Recognition. A report on the competition will be published in the proceedings of ASAR 2021. The report will comprise description of the participating methods, the evaluation protocol and the final rankings of the participating algorithms. The results of the competition will also be presented in a dedicated session at ASAR 2021.
REGISTRATION
Participants need to register in order participate in the competition. Please, fill up this form https://forms.gle/Ka2FyLvmhKCJFbon8 including your information.
Instructions
The competition aims to promote innovation in Online Arabic Words Recognition, as well as to provide objective and fair comparisons among methods. The ranking of participants is based on word recognition rate with respect to test samples. Participants will be given a limited time to submit their results after the competition started. The contest will consist of two phases:
-
Phase 1: Participants get training data extracted from ADAB Database to train their algorithms.
-
Phase 2: Registered participants (https://forms.gle/Ka2FyLvmhKCJFbon8) get the test data (without the ground-truth). They must submit their results within two weeks from the release of the test data with a short description of the approach used and the source code.
Data description
-
The training data used for this competition consists of 1011 online handwritten words of 100 classes extracted from ADAB Database (An inkml file including trajectory information),
-
The test data used for this competition consists of --- words.
-
The ADAB database (The Arabic handwriting Data Base) is developed in cooperation between the Institut fuer Nachrichtentechnik (IfN) and Research Groups in Intelligent Machines, University of Sfax, Tunisia. The text written is from 937 Tunisian town/village names advanced the research and development of Arabic on-line handwritten systems.
[1] H. Boubaker, A. Elbaati, N. Tagougui, M. Kherallah, H. Elabed, and A. M. Alimi, “Online Arabic Databases and Applications,” Book chapter in : Märgner, V., El Abed, H. (eds) “Guide to OCR for Arabic Scripts,” Chp. Part IV pp 541 – 557 , 2012 , Springer.
[2] M. Kherallah, N. Tagougui, A. M. Alimi, H. El Abed and V. Margner, “Online Arabic handwriting recognition competition”, In International Conference on Document Analysis and Recognition, pp. 1454-1458, 2011.
[3] H. El Abed, M. Kherallah, V. Margner and A. M. Alimi, “on-line Arabic handwriting recognition competition ADAB database and participating systems”, International Journal on Document Analysis and Recognition, volume 13, July 2010.
EVALUATION AND RANKING
The objective is to run each Arabic handwritten word recognizer (trained on the ADAB) on an already published part of the ADAB Database and on a new sample not yet published. The recognition results on word level of each system will be compared to correct recognized word. A dictionary can be used and have to include all 100 different words.
The system will be tested on a new dataset. Running a Recognizer:
-
your recognizer (called myrec) by invoking it from the command line as follows:
“myrec input.inkml output.txt …”
-
input.inkml: inkml file from the database of 100 classes of different words
-
output.txt: The output file should have one line for each input trace. Each response is the class of word or the correct word appears in the top 5 of results.
Note that the code will solely use by the Technical Committee to guarantee the fairness of the competition. We will not share this with any third party or make any commercial use of it.
IMPORTANT DATES
All deadlines are at 11:59 PM UTC. The competition organisers reserve the right to update the contest timeline if they see it necessary.
-
Training: April 5th, 2021.
-
Registration: April 5th, 2021 to May 31th, 2021.
-
Test sets available: June 1st, 2021.
-
Final submission deadline: June 15th, 2021.