Abstract: Provided is a method of synthesizing 3D joint data based on a multi-view RGB-D camera. The method of synthesizing 3D joint data includes: converting joint data collected from a depth camera of each of a plurality of RGB-D cameras from a depth camera coordinate system of each of the RGB-D cameras to a color camera coordinate system of each of the RGB-D cameras; calculating a confidence level of the joint data converted to the color camera coordinate system using a 3D joint recognition algorithm based on the joint data converted to the color camera coordinate system; applying a rotation matrix and a translation vector to the joint data and converting the joint data to a predetermined reference coordinate system; and obtaining a weighted-average of the joint data converted to the reference coordinate system using a weight calculated based on the confidence level to synthesize the joint data.

Abstract: A method of guaranteeing game quality by using an artificial intelligence (AI) agent is provided. The method includes extracting an item list (hereinafter referred to as an inspection item list) for inspecting quality of a target game, extracting and storing log data corresponding to a test performance result for each item of the inspection item list, performing imitation learning of an AI agent model on the basis of the stored log data, performing an automatic test for inspecting quality of the target game by using the AI agent model on which the imitation learning is completed, and automatically recording a bug and an error detected by the AI agent model.

Abstract: Provided are a context-aware model generating method, an important event determining method, and an in-game context management server for performing the same methods. The context-aware model generating method includes generating a context-aware model which determines whether an important event of a video game occurs by using in-game data, utterance data, and chat data collected from the video game capable of being broadcasted as a sports game video. The important event determining method includes combining candidate events predictable from each of the in-game data, the utterance data, and the chat data based on the context-aware model, and determining, as an important event, a candidate event including a duplicate game content among candidate events.

Abstract: Provided is a method of automatically creating an agent that may interact with a user in a video game and providing a pre-created agent upon request of the user playing the video game when there is no user to play the video game.

Abstract: According to an embodiment of the present disclosure, a method for learning an exercise posture of a user is disclosed. The method includes: checking joint feature point information which is constructed based on a joint of the user; learning a ready posture learning model by learning the joint feature point information corresponding to a ready posture of the user; and learning an exercise posture learning model by learning the joint feature point information corresponding to an exercise posture of the user.

Abstract: An apparatus and a method for predicting a result of a computer game at a specific time point using obtaining play attributes of the computer game in a time slot including the specific time point; and predicting the result of the computer game at the specific time point by inputting the play attributes to a game result prediction model corresponding to the time slot are provided.

Abstract: Provided is an impact motion recognition system for screen-based multi-sport coaching.

Abstract: Provided are an apparatus and method for experiencing an augmented reality (AR)-based screen sports match which enable even a child, an elderly person, and a person with a disability to easily and safely experience a ball sports match, such as tennis, badminton, or squash, as a screen sports match without using a wearable marker or sensor, a wearable display, an actual ball, and an actual tool.

Abstract: A device for generating a non-player character (NPC) capable of adjusting a skill level is provided. The device includes an NPC generator configured to generate an NPC having a skill level value adjusted based on a certain skill level adjustment parameter for each skill level factor, an NPC pool storage configured to store the generated NPC, and an NPC extractor configured to extract an NPC, corresponding to a skill level value input by a user, in the NPC pool storage and to provide the extracted NPC to content.

Abstract: A method of guaranteeing game quality by using an artificial intelligence (AI) agent is provided. The method includes extracting an item list (hereinafter referred to as an inspection item list) for inspecting quality of a target game, extracting and storing log data corresponding to a test performance result for each item of the inspection item list, performing imitation learning of an AI agent model on the basis of the stored log data, performing an automatic test for inspecting quality of the target game by using the AI agent model on which the imitation learning is completed, and automatically recording a bug and an error detected by the AI agent model.

Abstract: Provided are an apparatus and method for experiencing an augmented reality (AR)-based screen sports match which enable even a child, an elderly person, and a person with a disability to easily and safely experience a ball sports match, such as tennis, badminton, or squash, as a screen sports match without using a wearable marker or sensor, a wearable display, an actual ball, and an actual tool.

Abstract: An agent creating method of creating an agent in a game environment is provided. The agent creating method includes creating, by an agent creating apparatus, a base agent on the basis of a common action characteristic pattern of players to transfer the base agent to the game server, receiving, by the agent creating apparatus, match result data obtained by performing a match between the base agent and a character of an individual player, and performing, by the agent creating apparatus, machine learning by using the match result data and creating an evolved agent customized for matchmaking of the individual player on the basis of a result of the machine learning.

Abstract: An apparatus and a method for predicting a result of a computer game at a specific time point using obtaining play attributes of the computer game in a time slot including the specific time point; and predicting the result of the computer game at the specific time point by inputting the play attributes to a game result prediction model corresponding to the time slot are provided.

Abstract: Device for simulation environment for training Al agent includes scene object providing module to provide scene and object in virtual content converted from original content; a reward function providing module to provide reward function used by agent to perform reinforcement learning in the virtual content; an environment information providing module to provide virtual environment information including information on environment where the agent performs the reinforcement learning in the virtual content; a status information providing module to provide virtual status information indicating status of the agent in the virtual content; an action space providing module to provide virtual action space indicating action of the agent; and a virtual learning module to create simulation environment based on at least one of the scene, the object, the reward function, the virtual environment information, the virtual status information, and the virtual action space, and perform virtual learning for the agent in the simula

Abstract: The present invention provides a gamer behavior prediction modeling system based on time-series data conforming to an in-game behavior of a gamer and an environment attribute variation associated with a game in an online game service. Also, the present invention provides a method and system of generating a game operating scenario on the basis of a gamer behavior prediction model.

Abstract: Disclosed is an apparatus for analyzing a game update effect according to a change in a gamer action sequence, the apparatus including: a gamer action information collector configured to collect action information of a gamer from a game operating server that provides a gamer terminal with a game and stores the action information of the gamer therein; a gamer action information sequence identifier configured to detect an action sequence of the gamer from the collected action information of the gamer; and an update result analyzer configured to analyze an consequence on behavior of the gamer with respect to an update of a game service by comparing action sequences of the gamer detected through the gamer action information sequence identifier on the basis of a time point of the update of the game service.

Abstract: An agent creating method of creating an agent in a game environment is provided. The agent creating method includes creating, by an agent creating apparatus, a base agent on the basis of a common action characteristic pattern of players to transfer the base agent to the game server, receiving, by the agent creating apparatus, match result data obtained by performing a match between the base agent and a character of an individual player, and performing, by the agent creating apparatus, machine learning by using the match result data and creating an evolved agent customized for matchmaking of the individual player on the basis of a result of the machine learning.

Abstract: Disclosed is an apparatus for analyzing a game update effect according to a change in a gamer action sequence, the apparatus including: a gamer action information collector configured to collect action information of a gamer from a game operating server that provides a gamer terminal with a game and stores the action information of the gamer therein; a gamer action information sequence identifier configured to detect an action sequence of the gamer from the collected action information of the gamer; and an update result analyzer configured to analyze an consequence on behavior of the gamer with respect to an update of a game service by comparing action sequences of the gamer detected through the gamer action information sequence identifier on the basis of a time point of the update of the game service.

Abstract: The present invention relates to a spoken dialog system and method based on dual dialog management using a hierarchical dialog task library that may increase reutilization of dialog knowledge by constructing and packaging the dialog knowledge based on a task unit having a hierarchical structure, and may construct and process the dialog knowledge using a dialog plan scheme about relationship therebetween by classifying the dialog knowledge based on a task unit to make design of a dialog service convenient, which is different from an existing spoken dialog system in which it is difficult to reuse dialog knowledge since a large amount of construction costs and time is required.

Abstract: Disclosed herein is a game bot detection apparatus and method. The game bot detection apparatus delivers a natural language query to a game user and determines whether the game user uses a game bot, based on expression of a response to the natural language query, received from the game user so as to determine whether the game user uses the game bot. The game bot detection apparatus includes a query generation unit for generating a natural language query to be provided to the game user. A keyword extraction unit receives expression of a response to the natural language query, input by the game user, and extracts a keyword from the response expression. An association analysis unit analyzes an association between the natural language query and the keyword, and determines whether the game user uses a game bot, based on results of analysis of the association.