| Title |
Re-identification Model Pipeline for Daycare Children Behavior Monitoring |
| Authors |
조현인(Hyunin Cho) ; 최경묵(Kyungmook Choi) ; 어영정(Youngjung Uh) |
| DOI |
https://doi.org/10.5573/ieie.2024.61.11.129 |
| Keywords |
Object detection; Person re-identification; Spectral clustering |
| Abstract |
Teachers in daycare centers can manage only a limited number of children at a time, which creates challenges in ensuring the safety of children while monitoring their individual behaviors and daily routines. Recent advancements in artificial intelligence, especially in object detection, person re-identification, and object tracking, present new possibilities for addressing these challenges. This study proposes a method to automatically detect and identify children and teachers using CCTV video recordings in daycare centers. This approach aims to provide a technological foundation for preventing accidents and efficiently tracking daily activities of children, enabling quick evaluation of their behavior patterns and developmental progress. Furthermore, this study enhances practicality by presenting detection and identification methods suitable for real-world daycare settings as opposed to controlled laboratory environments. The results demonstrate the potential of this approach as an effective tool for safety management and child development monitoring in daycare contexts, contributing to technological innovation in early childhood education and care. |