AI-Powered Object Detection for Toddler Learning

This project introduces AI‑Powered Object Detection for Toddler Learning, a mobile application developed within the Kidzinya early‑learning initiative. It uses real‑time YOLO‑based object detection through a phone camera to support interactive learning for children ages 1–4. The system identifies objects and provides audio‑visual educational feedback to encourage play‑driven engagement. A custom dataset of 8,000+ images across 50+ toddler‑focused classes was collected, labeled, and curated for training. YOLOv8 was fine‑tuned using transfer learning, staged training, and class‑imbalance handling with Focal Loss. The app includes progress tracking, child profile management, and a privacy‑first on‑device processing design. Project materials cover objectives, key features, system requirements, and comparisons with similar educational apps. The literature review references YOLO‑based research in computer vision, early childhood education, indoor detection, and online safety. Documentation includes technologies used, functional and non‑functional requirements, diagrams, and a four‑week implementation timeline. This work was conducted at Arab International University (AIU), Syria. The official website of the university is: https://www.aiu.edu.sy