• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.333-336
• /
• 2003

컴퓨터와 네트워크의 처리속도 증가와, 인터넷의 발달로 인하여 이미지, 사운드, 동영상 등 각종 멀티미디어 정보가 인터넷상에 다수 등록되고 있으며, 이에 대한 검색 요구도 증가하고 있다. 그에 따라 다양한 멀티미디어 정보 검색을 위한 방법이 연구되고 있지만, 그에 대하 활용도는 미미하며, 데이터 베이스에 등록된 단순 멀티미디어 정보 검색에 머물고 있는 실정이다. 이에 본 연구는 인터넷상의 멀티미디어 정보 중 이미지 정보를 능동적으로 수집, 정보를 추출하여 검색에 이용한다 이를 위하여, 이미지에 대한 text 정보와 이미지의 속성 및 Link 정보를 이용, 의미 있는 이미지와 의미 없는 이미지를 분류하여 검색의 효율을 높이고, 속성 및 Link 정보를 가중치로 사용함으로써 검색 시 이미지의 중요도를 평가할 수 있도록 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.615-617
• /
• 2022

본 논문에서는 얼굴사진 기반 감정인식 심층망, 음성사운드를 기반한 감정인식 심층망을 결합한 앙상블 네트워크 구축을 위한 사전연구로서 얼굴사진 기반 감정을 인식하는 기존 딥뉴럴 네트워크 모델들을 입력 데이터 처리 방법에 따라 분류하고, 각 방법의 특성을 분석한다. 또한, 얼굴사진 외관 특성을 기반한 감정인식 네트워크를 여러 구조로 구성하고, 구성된 방법의 성능을 비교하여, 우수 성능을 보이는 네트워크를 선정하여 추후 앙상블 네트워크의 구성 네트워크로 사용하고자 한다.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.8
• /
• pp.481-491
• /
• 2005

The purpose of this study is to interpret the evaluation structure of auditory images about streetscapes in urban area on the basis of the semantic view of soundscapes. Using the caption evaluation method. which is a new method, from 2001 to 2005, a total of 45 college students participated in a fieldwork to find out the images of sounds while walking on the main streets of Namwon city. It was able get various data which include elements, features, impressions, and preferences about auditory scene. In Namwon city, the elements of the formation of auditory images are classified into natural sound and artificial sound which include machinery sounds, community sounds. and signal sounds. Also, the features of the auditory scene are classified by kind of sound, behavior, condition, character, relationship of circumference and image. Finally, the impression of auditory scene is classified into three categories, which are the emotions of humans, atmosphere of the streets, and the characteristics of the sound itself. From the relationship between auditory scene and estimation, the elements, features and impressions of auditory scene consist of the items which are positive, neutral, and negative images. Also, it was able to grasp the characteristics of auditory image of place or space through the evaluation model of streetscapes in Namwon city.

• The Journal of the Korea Contents Association
• /
• v.6 no.9
• /
• pp.108-115
• /
• 2006

This paper will study the influence of comedic elements of the game on the gaming choice by the game users, and focus its goal on establishing the foundation for the various game production environments. Within the game, the comedic elements are categorized as one of actively expressive elements and enhance the game's entertainment together with its sound qualities of graphics, scenario, sounds and level designs. Although the comedic elements are generally acknowledged as necessities, the research on how the users actually perceive those elements has been insufficient. Therefore this paper will investigate characteristics, compositions and techniques of the comedy used in the game and will analyze the influence that those comedic elements have on the users' recognition, satisfaction and royalty to the game.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.7
• /
• pp.1572-1580
• /
• 2010

This thesis presents an emotion-based contents player platform that can change its visual and aural components as user's emotions. It analyzes the emotion as pleasant, unpleasant, aroused, and relaxed based on the physiological signals and the user's active response. Accordingly. the system reorganizes graphical and aural stimuli, such as, light, color, sound, in real-time. It can be used to develop and show the emotional contents and also be applied for the systematic analysis to find out how the components would affect the emotion. This paper describes overall the system architecture and the implementations of the sub-systems, as well as the actual contents built on top of the platform.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.11a
• /
• pp.173-176
• /
• 2009

본 논문에서는 영상 정보와 오디오 정보 분석을 이용하여 TV 골프 방송 프로그램에서 중요 이벤트 구간을 검출하고 요약 하는 알고리즘을 제안한다. 제안하는 알고리즘에서는 입력되는 TV 골프 동영상을 영상 신호와 오디오 신호로 분리한 후에, 연속적인 오디오 스트림을 내용 기반의 오디오 구간으로 분류한 뒤 오디오 이벤트 구간을 검출하고, 이와 병렬적으로 영상정보에서 선수들의 플레이 장면을 검출한다. 플레이 장면 검출에 있어서는 방송 환경이나 날씨 등의 변화하는 다양한 조건에 대해 플레이 장면에 대한 오프라인 모델과 함께 경기 내에서 발생한 온라인 모델에 대한 학습을 혼합 적용함으로써 검출 성능을 높였다. 오디오 신호로부터 관중들의 박수소리와 스윙 사운드를 통해 검출된 오디오 이벤트와 플레이 장면은 이벤트 장면 검출 및 요약본 생성을 위해 사용된다. 제안된 알고리즘은 멀티 모달 정보를 이용하여 이벤트 구간 검출을 수행함으로써 중요 이벤트 구간 검출의 정확도를 높일 수 있었고, 검출된 이벤트 구간에 대한 요약본 생성을 통해 골프 경기를 시청하는 사용자가 원하는 부분을 빠르게 브라우징하여 시청하는 것이 가능하여 높은 사용자 만족도를 얻을 수 있었다.

• Journal of Korea Game Society
• /
• v.6 no.4
• /
• pp.89-95
• /
• 2006

The game has both directionalities that users are to choose and act accordingly. In this regard, the game scenario goes beyond just conveying the story, and trains users and guide them in playing the Same. However, this Same scenario is huge in size and every event is linked with each other so that it was rather difficult to have it made in practice. A study proposes that a script, the minimum unit, is built into a database, and based on which, quests are made, then the resulting quests are built back into a database. Scripts are classified into text, graphic and sound type and these are positioned in accordance with the quest structure. With this method, one can re-use the existing scenarios and can overcome the negatives of irregular qualities of scenarios.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.155-160
• /
• 2021

This paper proposes an automatic method for detecting snore sound, one of the important symptoms of sleep apnea patients. In the proposed method, sound signals generated during sleep are input to detect a sound generation section, and a spectrogram transformed from the detected sound section is applied to a classifier based on a Convolutional Bidirectional Gated Recurrent Unit (CBGRU) with attention mechanism. The applied attention mechanism improved the snoring sound detection performance by extending the CBGRU model to learn discriminative feature representation for the snoring detection. The experimental results show that the proposed snoring detection method improves the accuracy by approximately 3.1 % ~ 5.5 % than existing method.

• Journal of Digital Contents Society
• /
• v.13 no.1
• /
• pp.119-127
• /
• 2012

In this study, it was witnessed that a variety of tests are implemented in a type of convergence of multimedia such as photos, sounds, and videos as well as letters and drawings, components of existing traditional comics as comics are digitalized and are converted to various devices. Therefore, we studied the concept of multimedia comics as a basis of this study by judging that new barometer to comics lies in convergence with such multimedia. After recognizing components of multimedia comics which are currently emerging, we categorized them into three types depending on how to use these elements. First, convergence type webtoon has a very similar format with existing vertical scrolling webtoon and has characteristics that background & effects sounds are added to emphasize the features of webtoon, or photos or videos are inserted in part, and there is no function to control these elements; Second, motion comic, a medium format between comic and animation has a characteristic that sound, video, paging are auto-played like watching animation but it keeps the format of comics within one frame; Third, interactive comic has a characteristic that effects sound, motion, and story are made by active participation of viewers. As a result of analyzing comics which having above multimedia characteristics, its implications are as followings: First, multimedia elements should be used depending on genre, age, and media; Second, high level of control technology considering the features of comic-viewers is needed. In other words, in continuously evolving media environments, comic contents being proper to targets and use purposes of viewers should be developed. For this, multimedia elements of comics should be used in order that comic-viewers can have active & interactive communication with contents.

• The Journal of the Korea Contents Association
• /
• v.10 no.6
• /
• pp.286-296
• /
• 2010

The study of the development of DJ robots is to meet the demands of the music services which are changing very rapidly in the digital and network era. Existing studies, as a whole, develop music services on the premise of personalized environment and equipment, but the DJ robot is on the premise of the open space shared by the public. DJ robot gives priority to traditional space and music. Recently as the hospitality and demand for cultural contents of South Korea expand to worldwide, industrial use of the contents based on traditional or our unique characteristics is getting more and more. Meanwhile, the DJ robot is composed of a combination of two modules. One is to detect changes in the external environment and the other is to set the properties of the music by psychology, emotional engineering, etc. DJ robot detect the footprint of the temperature, humidity, illumination, wind, noise and other environmental factors measured, and will ensure the objectivity of the music source by repeated experiments and verification with human sensibility ergonomics based on Hevner Adjective Circle. DJ robot will change the soundscape of the traditional space being more beautiful and make the revival and prosperity of traditional music with the use of traditional music through BGM.