This study proposes an immersive storytelling reading platform integrating Generative AI and interactive Augmented Reality (AR) to enhance children’s reading engagement. The platform combines style-transfer-based object detection, LangChain-driven real-time story generation, and Unity-based AR interactions, offering a multi-sensory reading experience via hand tracking and dynamic text. Experiments showed a 21.6% improvement in object detection accuracy, and EEG results revealed increased theta/beta power and suppressed alpha power when using the proposed platform, indicating enhanced attention. These findings highlight the educational potential of interactive reading environments for the digital-native generation.

ABSTRACT
1. 서론
2. 연구 내용
    2.1 스타일 전이 기반 데이터 증강을 활용한 그림책 객체 탐지 모델 구축
    2.2 LangChain 기반 상호작용적 스토리텔링 파이프라인 설계
    2.3 Unity 기반 인터랙티브 AR 환경 구축 및 핸드트래킹 파이프라인 설계
    2.4 뇌파 기반 집중력 비교 실험 설계
3. 연구 결과
    3.1 그림책 이미지에 특화된 객체 탐지 성능 향상
    3.2 LangChain 기반 스토리텔링 콘텐츠 구현
    3.3 HandTracking 및 Dynamic Text를 활용한 인터랙티브 AR 콘텐츠 구현 결과
    3.4 EEG 밴드별 상대 파워 분석을 통한 집중도 비교
4. 논의
5. 결론
참고문헌

• Sung Eon Seok(Department of Computer Engineering, Tech University of Korea) | 성언석
• Dongyoung Seo(Department of Computer Engineering, Tech University of Korea) | 서동영
• Sungtaek Chung(Department of Computer Engineering, Tech University of Korea) | 정성택 Corresponding author
• Sang-Gyu Kim(Department of Bio Health Convergence Engineering, Tech University of Korea) | 김상규
• Miseon Shim(Department of Artificial Intelligence, Tech University of Korea) | 심미선