Das, Abhijit, Suwanwiwat, Hemmaphan, Pal, Umapada, and Blumenstein, Michael (2020) ICFHR 2020 Competition on Short answer ASsessment and Thai student SIGnature and Name COMponents Recognition and Verification (SASIGCOM 2020). In: Proceedings of the 17th International Conference on Frontiers in Handwriting Recognition. pp. 222-227. From: ICFHR 2020: 17th International Conference on Frontiers in Handwriting Recognition, 8-10 September 2020, Dortmund, Germany.

PDF (Accepted Author Version) - Accepted Version Restricted to Repository staff only

 

3

3

Abstract

This paper describes the results of the competition on Short answer ASsessment and Thai student SIGnature and Name COMponents Recognition and Verification (SASIGCOM 2020) in conjunction with the 17th International Conference on Frontiers in Handwriting Recognition (ICFHR 2020). The competition was aimed to automate the evaluation process short answer-based examination and record the development and gain attention to such system. The proposed competition contains three elements which are short answer assessment (recognition and marking the answers to short-answer questions derived from examination papers), student name components (first and last names) and signature verification and recognition. Signatures and name components data were collected from 100 volunteers. For the Thai signature dataset, there are 30 genuine signatures, 12 skilled and 12 simple forgeries for each writer. With Thai name components dataset, there are 30 genuine and 12 skilfully forged name components for each writer. There are 104 exam papers in the short answer assessment dataset, 52 of which were written with cursive handwriting; the rest of 52 papers were written with printed handwriting. The exam papers contain ten questions, and the answers to the questions were designed to be a few words per question. Three teams from distinguished labs submitted their systems. For short answer assessment, word spotting task was also performed. This paper analysed the results produced by their algorithms using a performance measure and defines a way forward for this subject of research. Both the datasets, along with some of the accompanying ground truth/baseline mask will be made freely available for research purposes via the TC10/TC11.

Actions (Repository Staff Only)

Item Control Page