Interactive System for Toddlers Using Doodle Recognition
No Thumbnail Available
Date
2024
Journal Title
Journal ISSN
Volume Title
Publisher
Springer Science and Business Media Deutschland GmbH
Abstract
Typing using the keyboard or using a mouse is hard for small children. In this paper, we proposed an interactive system to improve the learning ability of a toddler. The proposed doodle recognition system provides an attractive and efficient way to interact toddlers with computer systems by following the Human-Computer Interaction guidelines and deep learning. The most common practice that toddlers develop is scribbling random images, so we decided to use this skill to provide a gateway for the toddlers to interact thus and learning with computers by using our proposed simple interface. When the toddler (user) starts to scribble or draw something on the screen, whiteboard, or paper; the application goes into input mode, and as soon as the drawing is stopped the image on the screen or whiteboard is processed by the trained CNN model and the action is carried out based on the output of the model. © Springer Nature Switzerland AG 2024.
Description
Keywords
Convolutional Neural Network (CNN), Doodle recognition, Human-Computer Interaction (HCI), Learning ability, Toddlers
Citation
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2024, Vol.13102 LNCS, , p. 597-606
