• 제목/요약/키워드: Character Industry

검색결과 522건 처리시간 0.032초

캐릭터 웹드라마 제작을 위한 프로그램 개발 연구 (A study on program development for character web drama production)

  • 이현수;김민하;서지원;조성진;이종원;김정이
    • 문화기술의 융합
    • /
    • 제9권5호
    • /
    • pp.591-596
    • /
    • 2023
  • 본 연구는 10대 1인 미디어 제작자를 중점으로 하여 쉽고 편리하게 영상을 제작할 수 있는 프로그램을 개발하였다. 사용자 조사를 통해 나타난 10대 제작자들의 요구사항과 문제점을 파악하여 사용자가 원하는 캐릭터 커스터마이징 기능과 GPT를 활용한 감정 및 행동 추천 시스템을 구현하였다. 렌더링 과정에서는 OpenCV와 FFmpeg를 사용하여 오디오와 영상을 결합하여 최종 영상을 생성하였다. 영상 제작에 전문성이 없는 10대 사용자들은 간단한 인터페이스를 통해 웹드라마 캐릭터를 커스터마이징하고, GPT의 도움을 받아 감정과 행동을 추천받을 수 있다. 본 연구의 프로그램은 편집 및 연출에 전문성이 없는 10대 사용자들을 대상으로 양질의 영상 제작을 돕고, 영상 제작에 대한 진입장벽을 낮춰주며, 1인 미디어 산업의 발전에 기여할 수 있는 도구가 될 것으로 기대된다. 향후 모바일이나 세로해상도 버전의 영상을 고려한 영상 제작 환경을 제공할 수 있도록 할 예정이다.

캐릭터 웹드라마 요약 분석을 통한 간접광고 제품 추천 시스템 개발 (Recommendation System Development of Indirect Advertising Product through Summary Analysis of Character Web Drama)

  • 이현수;김정이
    • 한국인터넷방송통신학회논문지
    • /
    • 제23권6호
    • /
    • pp.15-20
    • /
    • 2023
  • 본 논문은 캐릭터 웹드라마에 적합한 간접광고 제품을 추천하는 인공지능(AI) 시스템 알고리즘 개발에 관한 연구이다. 본 연구는 웹드라마의 대사 작성에 있어 그에 어울리는 간접광고 제품을 추천함으로써 시청자의 콘텐츠 몰입도를 높이고, 드라마의 스토리를 보다 깊게 이해하는 데 도움을 주는 것을 목표로 한다. 본 연구에서는 자연어처리 모델 인 GPT를 활용하여 대사, 줄거리를 분석하고, 분석 결과를 바탕으로 소품형, 배경형 등 두 가지 유형의 간접광고 제품 추천 시스템을 개발한다. 이를 통해 웹드라마의 스토리에 부합하는 제품을 적절히 배치함으로써 간접광고가 자연스럽게 노출될 수 있도록 하고, 그로 인해 시청자들의 몰입도가 증가하며, 상품 홍보의 효과 또한 높인다. 숨겨진 뜻이나 문화적 뉘앙스를 완벽하게 이해하기 어려운 인공지능 모델의 한계와 학습에 필요한 충분한 데이터 확보가 어렵다는 한계가 있다. 그러나 본 연구는 AI가 창작물 제작에 어떻게 기여할 수 있는지에 대한 새로운 인사이트를 제공하고, 창의적 산업 분야에서 자연어 처리 모델의 활용 가능성을 넓히는 중요한 발판이 될 것이다.

프로젝트 상급자 특성이 프로젝트관리자의 보고 성향과 시스템 개발성과에 미치는 영향 (Effect of Senior Managers' Characteristics on Project Manager's Reporting Tendencies and Information Systems Development Performance)

  • 장시영;오만석
    • Journal of Information Technology Applications and Management
    • /
    • 제19권2호
    • /
    • pp.149-179
    • /
    • 2012
  • This study explores the reporting tendencies of project managers when sending a status reports to their respective senior managers in information systems development project. It also examines the effect of selective reporting on report quality and systems development performance. The research model is proposed based on literature survey as well as key points from Iacovou et al. [2009] which is an extension of selective reporting in information systems project. Analysis of responses from 257 project managers in information systems industry indicates that pessimistic reporting by project managers has negative effect on both report quality (mediator variable) and systems development performance, while optimistic reporting has no significant effect on systems development performance. The CEO interest has an impact on the senior manager's power, which in turn leads to optimistic reporting by project managers. The senior manager's character has an impact on the senior manager's leadership, which in turn leads to pessimistic reporting. No direct relationship exists either between CEO interest and optimistic reporting or between senior manager's character and pessimistic reporting. In addition, trust toward the senior managers is significantly related to pessimistic reporting. Finally, report tendencies are also found to be affected by the active/passive characteristics of project managers.

산업간호사의 업무수행과 직무만족도에 관한 연구 (A Study on the Satisfaction of Occupational Health Nursing Work Related Infuluencial in Rural Area)

  • 김성실;권동연
    • 한국보건간호학회지
    • /
    • 제12권1호
    • /
    • pp.37-54
    • /
    • 1998
  • For the purpose of investigating the job implementation of industrial health nurses catch hold of satisfaction, relational character of influence. A questionnaire survey was carried out on 131 nurses who employed at the manufacture of electronic products$(27.1\%)$, distribution industry$(38\%)$, service station$(22\%)$, construction$(10\%)$, et cetra The tool of this study was modified as completed from kang's(l991) checklist of questions were composed of the performance of nurses' duty(13 items), the factor of main influential character of nurses' job(11 items), job satisfaction(26 items). The results were as follows. 1. Industrial health nurses worked to labor follow up study after physical examination$(100\%)$, medication$(97.1\%)$, workers regular physical examination(95.5%), dressing$(94.1\%)$. 2. The factor of influence of job performance were personal relation$(100\%)$, satisfaction of nurses job performance$(86.7\%)$, environment of workplace$(86.7\%)$. 3. A period of long work experienced nurses felt the highest satisfaction as 3.44 point.

  • PDF

챗봇 사용 의도에 영향을 미치는 요인 탐색 - 금융 서비스에서의 챗봇 (Exploring Factors Influencing Usage Intention of Chatbot - Chatbot in Financial Service)

  • 이민규;박희준
    • 품질경영학회지
    • /
    • 제47권4호
    • /
    • pp.755-765
    • /
    • 2019
  • Purpose: Chatbots are widely diffusing across various industries to substitute human manpower in the industry. However as researchers only develop technology that is applied to chatbot, the diffusion is slow in progress. The purpose of this study is to propose useful implications to accelerate diffusion of chatbots across industries by analyzing the perception of customers. To achieve the research purpose this study analyzes causal effect relationship between characteristics of chatbot character, service quality, individual difference, and intention to use chatbot. Methods: This study developed a survey that contains various questionnaires for each construct based on literature review. Data collected through survey was tested for convergent validity and discriminant validity and further analyzed the relationship using PLS-SEM method to verify hypotheses. Results: Trustworthiness of the chatbot character, ease of use, application design, responsiveness, customization, assurance, inertia, and previous experience have significant influence on forming user satisfaction, consumer trust, and intention to use. The others, likability, appropriateness, technology anxiety, and need for interaction were not significant in this research. Conclusion: Although the constructs of the research model was significant in previous literatures, some do not have significant effect on intention to use chatbots. Based on the results, chatbot managers will be able to develop chatbot systems which are more appealing to users and more academic researchers will focus on analyzing user perception and intention.

가상 3D 패션 코디네이션 연구 (A Study of Virtual 3D Fashion Coordination)

  • 강인애;김효숙;최창석
    • 대한가정학회지
    • /
    • 제40권6호
    • /
    • pp.159-171
    • /
    • 2002
  • Today, many people seek for their own personal character which is distinguished from another people and they utilize fashion coordination as the was of expression their own image. In addition, interest in electronic commerce and cuber shopping mall on the internet is increasing. For this reason, visual and interesting virtual fashion coordination system is needed. The purpose of this study is to propose possibility of fashion coordination by virtual 3D model. For this study, 1. We make a 3D standard body model by automatic generation. 2. We make 3D fashion item (sleeveless top and flare skirt) by automatic generation. 3. We combine 3D body model with fashion item by special point, grouping and gap being between body and clothes. 4. We make textile palettes and textile DB for texture mapping and rendering. As a effect of this study, 1. It can give the chance to coordinate clothes suitable for their own character and bodyshape on the cuber space more speedily and variously. 2. It can help fashion internet shopping mall company can save a time, expenses and tries to advertise their new products, offer service for customers and lead customers to purchasing. 3. It can accumulate a database of design and textile for using by fashion and textile industry.

맥시멀리즘 특징을 이용한 패션디자인의 표현성 연구 (The Expression of Fashion Design Using a Maximalism Character)

  • 김현진;이은숙
    • 한국의류산업학회지
    • /
    • 제13권1호
    • /
    • pp.7-16
    • /
    • 2011
  • Maximalism is the code which has a power to interpret a fashion phenomenon expressed complexly and variously in the multicultural society. In the 21st century, maximalism which reappeared is not the expression by a method only. It is presenting the experimental design for the new pursuit through the mixture such as expansion, splendor, variety, decoration, distortion of configuration or transformation, and composition. Therefore, this study aims at analyzing expression of fashion design by four types(expansion, decoration, mixture, non-structure) materializing a feature of maximalism expressed in the domestic and foreign collection works from 2005 to 2010. The research results are as follows. 1. Expansion: It was showed artificial, formative, unique, and odd teatures through the expansion of the upper half of the body, the lower half of the body, and both of them. 2. Decoration: It was showed the actual decoration considering functionality and practicality and the decoration emphasizing and unique character and featuring domination and emphasis. 3. Mixture: It was showed through the mixture of the contrary textiles or the different sex image. 4. Non-structure: It showed the non-structure of the avant-garde trend and non-structure through decoration, detail, adjustment direction of a dress.

Development of a Low-cost Industrial OCR System with an End-to-end Deep Learning Technology

  • Subedi, Bharat;Yunusov, Jahongir;Gaybulayev, Abdulaziz;Kim, Tae-Hyong
    • 대한임베디드공학회논문지
    • /
    • 제15권2호
    • /
    • pp.51-60
    • /
    • 2020
  • Optical character recognition (OCR) has been studied for decades because it is very useful in a variety of places. Nowadays, OCR's performance has improved significantly due to outstanding deep learning technology. Thus, there is an increasing demand for commercial-grade but affordable OCR systems. We have developed a low-cost, high-performance OCR system for the industry with the cheapest embedded developer kit that supports GPU acceleration. To achieve high accuracy for industrial use on limited computing resources, we chose a state-of-the-art text recognition algorithm that uses an end-to-end deep learning network as a baseline model. The model was then improved by replacing the feature extraction network with the best one suited to our conditions. Among the various candidate networks, EfficientNet-B3 has shown the best performance: excellent recognition accuracy with relatively low memory consumption. Besides, we have optimized the model written in TensorFlow's Python API using TensorFlow-TensorRT integration and TensorFlow's C++ API, respectively.

브리티시니스 관점에서 살펴본 폴 스미스 패션 특성 (The Characteristics of Paul Smith Fashion from the View of Britishness)

  • 김소라;이금희
    • 패션비즈니스
    • /
    • 제24권1호
    • /
    • pp.43-59
    • /
    • 2020
  • The purpose of this study is to examine the design of Paul Smith based on characteristics of the Britishness. Britishness formed according to the national character to encompass the lifestyles and traits of individual people through a shared natural environment and historic experiences within the national boundaries of the U.K., amid current global trends toward gigantic single market. In every season, Paul Smith is presenting various authentic tailored suits, reflecting the empirical positivism, characterized by traditional conservatism through contemporary reinterpretation, characteristic individualism through a reproduction of the sub-culture, optimistic naturalism through fabric and printing on the basis of the Britishness. This study analyzed the characteristics of the Paul Smith brand distinguished through its unique highlights of strong symbolism of the well-balanced U.K., with very individualistic and innovative ideas presenting basic and methodical research of Britishness. Additionally, the results of this study are expected to facilitate in establishing the identity of the fashion brand and exploring its direction, and it will also provide a crucial basis to seek the methods to express the identity of Korean culture in this era of globalization, setting a new course for the fashion industry.

LOG함수의 특성을 이용한 영상잡음제거(1) (A Study on Image restoration Algorithm using LOG function character)

  • 권기홍
    • 한국컴퓨터산업학회논문지
    • /
    • 제6권3호
    • /
    • pp.447-456
    • /
    • 2005
  • 국부 반복 복원 처리는 영상 전체를 반복 복원하는 기존의 반복 복원과는 달리, 영상을 국부적으로 구분하여, 변화량이 큰 부분은 기존의 반복 복원으로 처리하고 변화량이 적은 부분은 LOG함수의 특성을 이용하여 신장 시킨 다음 처리하고, 다시 압축시키므로, 기존의 반복 복원 처리보다도 MSE(Mean Square Error)를 월등히 줄일 수 있을 뿐 아니라 변화량이 적은 부분도 처리가 잘되고, 또 기존의 반복처리 방법이 갖는, 적은 메모리 용량의 소요, 비선형 제약조건의 사용 가능, 약간의 변형으로 언제나 수렴성을 보장하는 등의 장점을 모두 가진다. 이 방법을 영상에 적용시킨 결과, MSE의 현저한 감소, 반복횟수 감소에 따른 반복시간 단축을 확인 할 수 있었다. 그러므로, 이 방법은 MSE를 줄이고 또한 처리 시간 단축을 목적으로 하는 영상의 복원에 적용될 수 있는 매우 우수한 방법임을 알 수 있다.

  • PDF