This page was machine-translated and may differ from the original. View original
Real-time game screen analysis...provides play strategies
Applying international e-sports competitions, overcoming dependence on game companies
Applying international e-sports competitions, overcoming dependence on game companies
A domestic research team applied an AI esports analysis platform that provides real-time winning rate prediction services to international esports competitions and received favorable responses.
The Electronics and Telecommunications Research Institute (ETRI) announced on the 5th that it has developed a technology that recognizes real-time game situations by analyzing play elements extracted from game videos, recognizes key play events in the game, and automatically generates highlights. This was also applied to the recently held international e-sports competition game 'League of Legends (LoL)' and received a favorable response.
The platform records gamer profiles from game play data and provides AI-based play strategies. ETRI said, “This platform is the first in the world to provide real-time service for international e-sports competitions, and has created a prediction model that takes into account game time zone characteristics, securing a prediction accuracy of over 87%.”
Existing services were limited to commentary-based broadcasts due to the difficulty in securing access to game publisher APIs, but with this development, we have developed a technology that supports not only key play indicators but also various prediction information through real-time game screen analysis.
ETRI said, “By overcoming dependence on existing game companies, it can be applied to various game genres, which will be of great help in creating and commercializing new services.”
The e-sports service platform developed by ETRI is a SW technology that supports real-time services for various game genres, including △real-time game situation recognition, △automatic highlight creation, △gamer profile creation, and △play strategy recommendation.
The research team also developed an automatic game highlight generation service using in-game major event recognition technology. Unlike existing services, this service can be applied in various ways as it allows users to select highlight content based on their preferences. The service is scheduled to be launched on the market soon.
ETRI has developed a technology that analyzes large-scale play data to create profiles for individual gamers as well as teams. It is expected to recommend a precise and systematic training program and even provide multi-faceted play strategies.
This technology was transferred to Loud Corporation and provided for the 15th Presidential Amateur E-Sports Competition held last year, the 2023 E-Sports University League, and the 2023 Korea-China-Japan E-Sports Competition.
In the future, ETRI also plans to develop an automation platform for e-sports service operations to support small and medium-sized enterprises and regional e-sports stadium operators who are having difficulties securing human resources and developing new services.
Kim Kyung-hwa, head of the Cultural Industry Policy Division at the Ministry of Culture, Sports and Tourism, said, “As the global e-sports industry is growing significantly and its influence on other industries is increasing, we plan to actively support securing industry leadership through discovering creative service models through the convergence of new technologies.”
ETRI Content Research Center Director Jeong Il-kwon also said, “The real-time winning rate prediction technology applied to the 2023 Korea-China-Japan e-sports competition operates without API support from game companies, so it has excellent scalability. In addition, we plan to expand the application to various genres of e-sports to provide an enjoyable viewing experience for spectators and contribute to the revitalization of e-sports broadcasting services.”
본 기사에 대한 정정·반론·추후보도 청구는 보도 청구 안내를, 그간 게재된 보도문은 정정·반론보도 모아보기를 참고해 주세요.















