• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.227-241
• /
• 2021

Object detection techniques based on deep learning such as YOLO have high detection performance and precision in a single channel video stream. In order to expand to multiple channel object detection in real-time, however, high-performance hardware is required. In this paper, we propose a novel back-end server framework, a real-time AI vision platform (RAVIP), which can extend the object detection function from single channel to simultaneous multi-channels, which can work well even in low-end server hardware. RAVIP assembles appropriate component modules from the RODEM (real-time object detection module) Base to create per-channel instances for each channel, enabling efficient parallelization of object detection instances on limited hardware resources through continuous monitoring with respect to resource utilization. Through practical experiments, RAVIP shows that it is possible to optimize CPU, GPU, and memory utilization while performing object detection service in a multi-channel situation. In addition, it has been proven that RAVIP can provide object detection services with 25 FPS for all 16 channels at the same time.

• Journal of Internet Computing and Services
• /
• v.23 no.1
• /
• pp.69-76
• /
• 2022

Education systems have made efforts to prepare students by providing technical and nontechnical courses. With video games, however, there is the potential to develop dedicated metaverses that can help teach soft skills even during casual pastimes. The research conducted will propose a set of design practices for metaverse and game development to promote soft skills. While there are many soft skills people can acquire, this paper will focus on certain aspects based on specific games and studies. There will be some information collected from the information to support the design model and arguments. This paper will provide developers with a starting point for imaginative game creation and impart users with soft skills to assist in their professions and social life.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.4
• /
• pp.11-19
• /
• 2022

Data mining has broad applications that reach beyond scholarly and scientific research and provide internet search engine services that are commonly used forms of Text and Data Mining('TDM') of websites. The exceptions and limitations for data mining provide a competitive advantage in the global race for policy innovation because it permits researchers to conduct computational analysis - TDM on any materials to which they have access. For this purpose, Japan and the EU added limitations on copyright to legalize some TDM research through amendments to copyright law, and the U.S. copyright law has allowed data mining by the fair use provision. On the other hand, there are no explicit exceptions and limitations for data mining under the Korean Copyright Act, and there are no cases considering data mining fair use. We review comparatively exceptions and limitations on copyright which will help to encourage AI-related business by using more data smoothly through the mining process and extracting more valuable information.

• The Journal of Information Systems
• /
• v.31 no.2
• /
• pp.137-158
• /
• 2022

Purpose As artificial intelligence voice agents (AIVA) have been widely adopted in services, diverse forms of their voices, which are the main interface with users, have been experimented. The purpose of this study is to examine how users evaluate vocal characteristics (gender, voice pitch, and voice pace) of AIVA, depending on prior research on human voice attractiveness. Design/methodology/approach This study employed an experimental survey which 516 participated in. Each participant was randomly assigned into one of eight situations (e.g., male - higher pitch - faster pace) and listened a AIVA voice sample, which introduce weather information. Next, a participant answered three consequence factors (attractiveness, trust, and anthropomorphism). Findings The results reveal that female voices of AIVA were perceived as more attractive and trustworthy than male voices. As far as voice pitch goes, while lower-pitch voices were preferred in female voices, higher-pitch voices were preferred in male voices. Finally, faster voices of AIVA were more attractive than slower voices.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.619-623
• /
• 2022

대화생성은 대규모 학습 데이터로부터 사전 학습된 언어모델을 활용한 도전적인 다운스트림 태스크 중 하나이다. 대화에서 특정한 지식에 대한 맥락이 보존된 응답 문장을 생성하기 위한 기술의 일환으로써 지식 기반 대화생성이 연구되고 있으며, 현업에서는 사업목표에 따른 대화 서비스를 제공하는 목적으로 이러한 기술을 적용할 수 있다. 본 논문에서는, 각각의 서비스 도메인에 특화된 모델을 적절히 활용 가능하도록 전체 데이터를 도메인별로 구분하여 학습한 다수의 대화생성 모델을 구축한다. 또한, 특정 도메인의 데이터로 학습된 모델이 나머지 도메인에서 어떤 수준의 대화생성이 가능한지 비교 분석함으로써 개별 학습된 모델들이 도메인의 특성에 따라 서로 다른 영향력이나 연관성을 나타낼 가능성을 확인한다. 이러한 실험적인 분석 결과를 바탕으로 현업의 서비스에서 개별 도메인에 특화된 모델이 적절히 활용하는 것이 유용함을 확인하고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.575-578
• /
• 2020

최근 교육부는 초등학교부터 중학교까지 정보통신기술(ICT)와 인공지능(AI)에 관한 소양을 길러주기 위해 AI, VR 등 최첨단 기술을 적용한 '지능형 과학실'을 2024년까지 모든 학교에 구축할 방침이라고 한다. 하지만 국내 VR 교육은 학년별, 교과과정에 맞춘 콘텐츠가 부족하고, VR 교육 전용 LMS(학습관리 시스템)의 부재로 현실로 도입하기에 부족하다. 본 논문에서는 VR 교육 특성에 맞는 LMS 대안과 10분 내외의 VR 체험을 뒷받침할 맞춤 콘텐츠로서 'Web & VR Hybrid Content'를 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.711-713
• /
• 2023

딥러닝 기반 성별 및 연령 추정은 현재 다양한 응용분야에 적용하기 위하여 활발히 연구되어 왔다. 최근 몇몇 기업에서는 개인정보 보호의 문제로 성별 추정 서비스를 중단하고 있기는 하나, 사용자 및 관람객의 특성 분석등을 위하여 그 필요성이 계속하여 대두되고 있는 상황이다. 본 논문은 기존에 성별 및 나이 추정을 위한 유료서비스에 대한 특성과 사용 문제점 등을 세분화하여 조사하고 성별 및 연령 추정 기반 기존 연구들의 특성을 분석한다. 추후 이 서비스를 기반으로 전시 · 컨벤션 산업에서의 관심도 분석 및 재방문 의사를 파악하는데 활용하고자 한다.

• Journal of Engineering Education Research
• /
• v.26 no.6
• /
• pp.71-78
• /
• 2023

ChatGPT significantly broadens the application of artificial intelligence (AI) services across various domains, with one of its primary functions being assistance in programming and coding. Nevertheless, due to the short history of ChatGPT, there have been few studies analyzing its coding capabilities in Korean higher education. In this paper, we evaluate it using exam questions from three foundational programming courses at S University. According to the experimental results, ChatGPT successfully generated Python, C, and JAVA programs, and the code quality is on par with that of high-achieving students. The powerful coding capabilities of ChatGPT imply the need for a strict prohibition of its usage in coding tests; however, it also suggests significant potential for enhancing practical exercises in the educational aspect.

• International conference on construction engineering and project management
• /
• 2011.02a
• /
• pp.597-602
• /
• 2011

A robust asset management methodology is essential for effective decision-making of maintenance, repair and rehabilitation of a bridge network. It can be achieved by a computer-based bridge management system (BMS). Successful BMS development requires a reliable bridge deterioration model, which is the most crucial component in a BMS, and an optimal management philosophy. The maintenance optimization methodology proposed in this paper is developed for a small bridge network with limited structural condition rating records. . The methodology is organized in three major components: (1) bridge health index (BHI); (2) maintenance and budget optimization; and (3) reliable Artificial Intelligence (AI) based bridge deterioration model. The outcomes of the paper will help to identify BMS implementation problems and to provide appropriate solutions for managing small bridge networks.

• Science of Emotion and Sensibility
• /
• v.22 no.3
• /
• pp.35-46
• /
• 2019

Artificial intelligence (AI) technology has been evolving to recognize and learn the languages, voice tones, and facial expressions of users so that they can respond to users' emotions in various contexts. Many AI-based services of particular importance in communications with users provide emotional interaction. However, research on nonverbal interaction as a means of expressing emotion in the AI system is still insufficient. We studied the effect of lighting on users' emotional interaction with an AI device, focusing on color and flickering motion. The AI device used in this study expresses emotions with six colors of light (red, yellow, green, blue, purple, and white) and with a three-level flickering effect (high, middle, and low velocity). We studied the responses of 50 men and women in their 20s and 30s to the emotions expressed by the light colors and flickering effects of the AI device. We found that each light color represented an emotion that was largely similar to the user's emotional image shown in a previous color-sensibility study. The rate of flickering of the lights produced changes in emotional arousal and balance. The change in arousal patterns produced similar intensities of all colors. On the other hand, changes in balance patterns were somewhat related to the emotional image in the previous color-sensibility study, but the colors were different. As AI systems and devices are becoming more diverse, our findings are expected to contribute to designing the users emotional with AI devices through lighting.