IMPLEMENTASI KOMPUTER VISION PADA GAME EDUKASI BANGUN RUANG UNTUK ANAK USIA DINI

Abstract

Learning two-dimensional shapes in early childhood education often faces challenges due to the limitations of conventional learning media, which are static in nature. Teachers typically use physical aids such as wooden puzzle pieces, but these materials are easily damaged and often lose parts. This study proposes the development of an educational game based on computer vision that enables interaction through real-time hand movements, making the learning process more interactive and engaging. The research method uses the System Development Life Cycle (SDLC), which includes needs analysis, system design, implementation using Python, OpenCV, MediaPipe, and Pygame, testing, and evaluation. The system is designed to recognize users’ hand movements and allow them to move two-dimensional shape objects to the designated target area (drop zone). Testing conducted with students at TK Tunas Satria shows that the game improves understanding of two-dimensional shape concepts, achieving an average score of 4.3 out of 5. The engaging interface and gesture-based interaction help students more easily recognize and differentiate geometric shapes. The main challenges lie in the need for good lighting conditions and specific hardware requirements. The findings demonstrate that the game effectively serves as an interactive learning tool, receiving positive reception from both children and educators. The integrated hand tracking technology successfully facilitates intuitive and captivating user interactions.