Name
Abstract | ||
|---|---|---|
This paper proposes a new playful approach to improve children's writing skills in primary school. This approach is currently being developed in the Study project (I-Study Project, French Region Auvergne-Rhone-Alpes R&D Booster Program, 2020-2022), that proposes a global learning solution through the use of an innovative interactive device, combining advanced technologies for the writing acquisition (Advanced Magnetic Interaction tech from AMI/ISKN 2 (ISKN Repaper Tablet,littps://v,vv and the conception of a gamified environment for children (Super Extra Lab 3 (SuperExtraLab, http://ww, superextralab.com/) distributed by Extrapage, publisher of the solution) 4 (Extrapage, https://www.extrapage.io/fr/pages/index.html) The proposition associates traditional school materials (textbook, paper, pencil...) with very recent technologies and didactic innovations. While the state-of-art is usually satisfied with solutions strictly focusing on lexical spelling, the Study project aims to contribute to a broader control of handwriting skills, i.e., the acquisition of efficient handwriting micromotor skills and also strong grammatical skills. To achieve this, the approach relies on a personalized, regular and enriched copy and dictation training program. It establishes conditions that encourage children to be involved in an activity that is often unpleasant, and is inspired by the mechanics of video games. In the paper, we present an overview of the project, the construction of the general framework in relation with the design of a game interface for learning and the first recognition results based on a dedicated MDSLTM-CTC trained on a GANgenerated dataset that cover different writing styles. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1007/978-3-030-86159-9_19 | DOCUMENT ANALYSIS AND RECOGNITION, ICDAR 2021, PT II |
Keywords | DocType | Volume |
Child handwriting recognition, Dictation, French language, ISKN Repaper, Extrapage publisher, SCOLEDIT dataset Study R&D project, MDLSTM technologies, GAN-generated handwriting styles, Human computer interaction, Deep-learning | Conference | 12917 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sofiane Medjram | 1 | 0 | 0.34 |
| Véronique Eglin | 2 | 131 | 19.67 |
| Stéphane Bres | 3 | 127 | 14.42 |
| Adrien Piffaretti | 4 | 0 | 0.34 |
| Jobert Timothée | 5 | 0 | 0.34 |