대화형 유전자 알고리즘을 이용한 감성기반 비디오 장면 검색 = Emotion - based Video Scene Retrieval using Interactive Genetic Algorithm

국문 초록 (Abstract)

본 논문에서는 감성에 기반한 장면단위 비디오 검색방법을 제안한다. 먼저 특정 줄거리를 담은 장면 비디오 클립에서 급진적/점진적 샷 경계 검출 후, “평균 색상 히스토그램”, “평균 밝기”, “평균 에지 히스토그램”, “평균 샷 시간”, “점진적 샷 변화율”의 5가지 특징을 추출하고, 이 특징과 사람이 막연하게 가지고 있는 감성공간과의 매핑을 대화형 유전자 알고리즘(IGA, Interactive Genetic Algorithm)을 통하여 실현한다.
제안된 검색 알고리즘은 초기 모집단 비디오들에 대해 찾고자 하는 감성을 내포하고 있는 비디오를 선택하면 선택된 비디오들에서 추출된 특징 벡터를 염색체로 간주하고 이에 대해 교차연산(crossover)을 적용한다. 다음에 새롭게 생성된 염색체들과 특징벡터로 색인된 데이타베이스 비디오들간에 유사도 함수에 의해 가장 유사한 비디오들을 검색하여 다음 세대의 집단으로 제시한다. 이와 같은 과정을 여러 세대에 걸쳐서 실행하여 사용자가 가지고 있는 감성을 내포하는 비디오 집단들을 얻게 된다. 제안된 방법의 효과성을 보이기 위해, 300개의 광고 비디오 클립들에 대해 “action”, “excitement”, “suspense”, “quietness”, “relaxation”, “happiness” 의 감성을 가진 비디오를 검색한 결과 평균 70%의 만족도를 얻을 수 있었다.

다국어 초록 (Multilingual Abstract)

An emotion-based video scene retrieval algorithm is proposed in this paper. First, abrupt/gradual shot boundaries are detected in the video clip representing a specific story. Then, five video features such as “average color histogram”, “average brightness”, “average edge histogram”, “average shot duration”, and “gradual change rate” are extracted from each of the videos and mapping between these features and the emotional space that user has in mind is achieved by an interactive genetic algorithm.
Once the proposed algorithm has selected videos that contain the corresponding emotion from initial population of videos, feature vectors from the selected videos are regarded as chromosomes and a genetic crossover is applied over them. Next, new chromosomes after crossover and feature vectors in the database videos are compared based on the similarity function to obtain the most similar videos as solutions of the next generation. By iterating above procedures, new population of videos that user has in mind are retrieved. In order to show the validity of the proposed method, six example categories such as “action“, “excitement”, “suspense”, “quietness”, “relaxation”, “happiness” are used as emotions for experiments. Over 300 commercial videos, retrieval results show 70% effectiveness in average.

목차 (Table of Contents)

• 요약
• Abstract
• 1. 서론
• 2. 비디오 샷 경계 검출
• 3. 특징치 추출
• 4. 감성기반 비디오 검색
• 5. 실험
• 6. 결론
• 참고문헌
• 저자소개

참고문헌 (Reference)

1 "VisualSEEK: A Fully Automated Content-Based Image Query System" 87-98, 1996

2 "Visual Information Retrieval System via Content-Based Approach" 35 (35): 749-769, 2002

3 "Tracking a Criminal Suspect through Face-Space with a Genetic Algorithm" 416-421, 1991.

4 "Towards Creative Evolutionary Systems with Interactive Genetic Algorithm" 16 (16): 129 2002-138,

5 "The Bayesian Image Retrieval System, PicHunter : Theory, Implementation and Psycophysical Experiments" 9 (9): 20-37, 2000.

6 "Semantics in Visual Information Retrieval" 6 (6): 38-53, 1999.

7 "Psychological Space to Hold Impression among Media in Common for Media Database Retrieval System" 263-268, 1999.

8 "Performance Characterization of Video-Shot- Change Detection Methods" 10 (10): 1-13, 2000.

9 "On Image Classification: City Images vs. Landscapes" 31 (31): 1921-1936, pp.1921-19361998.

10 "Objective Evaluation of Color Design" 12 (12): 184-194, 1987.

11 "Interactive Learning Using a Society of Models" 30 (30): 565-581, 1997

12 "Interactive Genetic Algorithm for Content-Based Image Retrieval" 479-484, 1998.

13 "Interactive Evolution" 1997.

14 "Image Classification for Content- based Indexing IEEE Trans. on Image Processing" 10 (10): 117-130, 2001.

15 "I"Blobworld: Image Segmentation Using Expectation-Maximization and Its Application to Image Querying" 24 (24): 1026-1038, 2002

16 "Genetic Algorithms in Search" 1989.

17 "Fusion of the Capabilities of EC Optimization and Human Evaluation," 89 (89): 1275-1296, 20002001

18 "Fusion of the Capabilities of EC Optimization and Human Evaluation Proc. of the IEEE" 89 (89): 1275-1296, 2001.

19 "Extraction of Major Object Features Using VQ Clustering for Content-Based Image Retrieval" 35 (35): 1115-1126, 2002

20 "Automated Video Segmentation Using Computer Vision Technique" 2 (2): 2003

21 "Art of Color" 1961

22 "AThe Virage Image Search Engine: An Open Framework for Image Management,"" 76-86, 1996

23 "A"Relevance Feedback: A Power Tool in Interactive Content-Based Image Retrieval" 8 (8): 644-655, 1998.

24 "A Toolbox for Navigating Large Image Databases" 7 (7): 184-198, 1999

25 "A Study of the Emotional Evaluation Models of Color Patterns Based on the Adaptive Fuzzy System and the Neural Network" 27 (27): 208-216, 2002.

26 "A Genetic Algorithm for Generating Jazz Solos" 131-137, 1994.

27 ""Retrieval of Commercials by Semantic Content: The Semiotic Perspective"" 13 (13): 93-118, 2001.

28 ""Query by Image Content: The QBIC System" 28 (28): 23-31, 1995

29 ""Photobook: Content-Based Manipulation of Image Databases,"" 18 (18): 233-254, 1996