- Development of On-Device Generative AI Video Optimization Technology… Fostering AI Talent Based on Industry-Academic Collaboration
Gachon University (President Lee Gil-yeo) is participating as a joint research and development institution in the ‘Generative AI Leading Talent Development Project’ promoted by the Ministry of Science and ICT and the Institute of Information and Communication Technology Planning and Evaluation (IITP), and is embarking on industry-academia cooperation and the cultivation of specialized talent in the field of generative AI.
This project is a national strategic initiative designed to foster master’s and doctoral-level professionals by linking the development of core generative AI technologies with practical education. It will be implemented over approximately 3 years and 9 months from this year until 2029 with a total budget of 7.25 billion won, and companies and universities will jointly conduct research on on-device, multimodal, and LLM-based generative AI technologies and operate educational courses.
In this leading talent development project for Generative AI, Persona AI serves as the lead organization, with Gachon University, Korea University, and Chung-Ang University participating as joint R&D institutions to carry out the ‘Development of a High-Performance, Lightweight H-MoE-Based On-Local Multi-Domain Generative AI System.’
A research team from Gachon University is undertaking the project titled “Development of On-Device Lightweight Generative AI Video Optimization Technology Based on Display Characteristics” in this initiative. The core of the project is the development of technology that automatically optimizes generative AI video quality and output methods by analyzing the output environment in real time, including display size, aspect ratio, and curvature.
The research involves Professor Kim Jeong-yun, the principal investigator of the sub-project, as well as Professors Jeong Yun-hyeon and Choi Jae-yong from the Department of Artificial Intelligence and Professor Lee Su-hyeon from the Department of Computer Engineering. Leveraging their experience in convergence research across AI, XR, and digital content, the research team plans to enhance the industrial applicability of generative AI technology and expand its scope of application to immersive media, XR content, and digital services in the public and medical sectors.
Professor Kim Jeong-yun stated, “We plan to secure core technologies and establish a demonstration system so that generative AI technology can be stably implemented in various device environments,” adding, “By linking research and development with education, we will create a model for cultivating convergent AI talent applicable in industrial settings.”
