• The Journal of the Acoustical Society of Korea
• /
• 제18권4E호
• /
• pp.43-50
• /
• 1999

A fast converging coding algorithm that adaptively selects the modes of macroblocks is introduced. For a given frame, the optimal modes are selected based on the decision curves that minimize the overall distortion at a given bit rate. The method proposed in this paper is different from the conventional ones in that it does not manipulate the quantizer to meet the target bit rate but it satisfies the target bit rate by finding optimal modes of macroblocks which result consistent visual quality. Lagrange multiplier of the unconstrained cost function is controlled to trigger decision curves to generate appropriate modes to meet bit rate and the curve is obtained by utilizing simulated annealing optimization technique. The algorithm is implemented within H.261 video codec and simulation results demonstrate superior visual quality.

• 전자공학회논문지B
• /
• 제31B권10호
• /
• pp.130-139
• /
• 1994

Video compression algorithms are based on removing spatial and/or temproal redundancy inherent in image sequences by predictive(DPCM) encoding, transform encoding, or a combination of predictive and transform encoding. In this paper, each 8$\times$8 DCT coefficient of DFD(displaced frame difference) is adaptively quantized by one of the four quantizers depending on total distortion level, which is determined by characteristics of HVS(human visual system) and buffer status. Therefore, the number of possible quantizer selection vectors(patterns) is 4$^{64}$. If this vectors are coded, toomany bits are required. Thus, the quantizer selection vectors are limited to 2048 for Y and 512 for each U, V by the proposed method using SWAD(sum of weighted absolute difference) for discriminating vectors. The computer simulation results, using the codebook vectors which are made by the proposed method, show that the subjective and objective image quality (PSNR) are goor with the limited bit allocation. (17Mbps)

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계학술대회논문집
• /
• pp.682-685
• /
• 2005

In this research, how the evaluation of the spacial image influenced by the environmental friendly elements included in the visual information, and how the selection of the sound changed depending on the characteristics of spatial image by the 40 subjects were carried out. Vast tracts of green land and the waterfront were highly preferred and impressive than the other spaces. The green music, signal with water sound and bird chirping sound were highly scored. In the frequency characteristics of the factors, the first factor was artificial sound(high at the low frequency band), the second was natural sound(uniform at all frequency band) and the third was water sound(high at the middle and high frequency band over 500Hz). This shows that the proposal of the sound which has the frequency characteristics fit to the spacial image should be selected for the soundscape of the target space.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제27권5호
• /
• pp.510-520
• /
• 2000

내용 기반의 비디오 인덱싱 및 검색을 위해서는 비디오 데이타를 셧(shot)으로 분할하고, 또 각 셧을 나타내는 대표 프레임을 선택하는 것이 필요하다. 하지만, 대표 프레임을 선택하는 것은 주관적이어서 일관되게 자동적으로 대표 프레임을 선택하는 것은 쉬운 문제가 아니다. 본 논문에서는 각 프레임에서의 영역을 바탕으로한 컨텐트 정보 및 시간 축 상의 변화를 이용하여 계층적으로 대표 프레임을 선택하는 방법을 제안한다. 먼저, 비디오 셧에서 카메라 모션을 검출하여 이에 따라 비디오 셧을 분류한다. 다음, 분류된 비디오 셧에 컨텐트의 중요도를 계산하기 위한 퍼지 규칙을 적용하여 대표 프레임을 선택한다. 끝으로, 선택되는 대표 프레임의 수는 브라우징 상세도(detailness)에 따라 계층적으로 선택되게끔 한다.

• International Journal of Contents
• /
• 제11권3호
• /
• pp.47-53
• /
• 2015

Visual object tracking is a fundamental problem in the field of computer vision, as it needs a proper model to account for drastic appearance changes that are caused by shape, textural, and illumination variations. In this paper, we propose a feature-based visual-object-tracking method with a sparse representation. Generally, most appearance-based models use the gray-scale pixel values of the input image, but this might be insufficient for a description of the target object under a variety of conditions. To obtain the proper information regarding the target object, the following combination of features has been exploited as a corresponding representation: First, the features of the target templates are extracted by using the HOG (histogram of gradient) and LBPs (local binary patterns); secondly, a feature-based sparsity is attained by solving the minimization problems, whereby the target object is represented by the selection of the minimum reconstruction error. The strengths of both features are exploited to enhance the overall performance of the tracker; furthermore, the proposed method is integrated with the particle-filter framework and achieves a promising result in terms of challenging tracking videos.

• 한국의류학회지
• /
• 제16권3호
• /
• pp.257-269
• /
• 1992

In the area of fashion business, market segmentation strategy has been paid attention for the purpose of assigning proficiently marketing resources. The use of fashion information in purchase decision process can serve as a base for market segmentation strategies. The purpose of this study was to identify four segmented profiles which are labelled as Print-oriented, Audio-visual oriented, Store intensive, and Pal advice group. Objectives were to determine fashion interest, fashion attitude and apparel selection criteria from each segment. For this study, the questionnaire was adminisered to 261 teachers and data were analyzed by using ANOVA, Regression and Pearson's Correlations. The results were as follows. 1. Print-oriented group had a positive attitude about fashion and they tended to place value on aesthetic and other-oriented criteria in selecting apparel. 2. Audio-visual oriented group had a strongly positive attitude about fashion but they tended to be less active toward buying fashion products and place much more value on other-oriented criteria in selecting apparel. 3. Store intensive group tended to be active toward buying fashion products. They had a positive attitude about fashion and placed value on aesthetic criteria in selecting apparel. 4. Pal advice group had a less positive attitude about fashion and tended to place value on economical and practical criteria in selecting apparel. Based on the profile of the each segment, it was suggested to express lingual or non-lingual symbols and to create concepts for meeting the needs of the segment

• 한국멀티미디어학회논문지
• /
• 제19권6호
• /
• pp.1074-1088
• /
• 2016

Successful Visual Story Telling(VST) of desire is a crucial key for the success of animation because desire is the leading power of story development of animation. An analysis of the desire of VST using the top 5 successful American feature film animations is carried out. Totally, 147 desire shots are extracted by using the proposed Objective Selection of Desire Shots(OSDS) method based on the theory of Makee's conflict and desire pursuing modeling, Maslow's 20 desire types, Greimas's actant model, and the 17 narrative process classification. In addition to them, the 5 Beat(5B) model of a scene is proposed. Five image specialists have evaluated VST of the selected 147 desire shots. For each shot, the desire type among the 20 desires and the strength are obtained. Among them, the top 3 desires(affiliation, achievement, and nurturance) appearing 51.8% are analyzed. The composition elements of shots affecting the desire type and the strength have found. These can be used for better VST of preproduction and production of animation.

• 한국의상디자인학회지
• /
• 제22권1호
• /
• pp.97-111
• /
• 2020

Classifying clothing fabrics into fashion images lacks research, but is necessary due to the short cycle of fashion and rapidly changing modern trends as consumers seek to satisfy various needs with an increase of online purchases. There is also a lack of research on fashion trend books which attach real fabrics. Therefore, this study aims to help the planning field by recognizing fabric sensibility related to the fabric image perceived by consumers. The data analysis results from descriptive statistics, t-test, and ANOVA using SPSS are as follows: Differences in the visual tactility evaluation related to the consumer recognized fabric images showed more significant differences in F/W seasons. The elegance image was shown as relatively thick, the avant-garde image was shown as relatively heavy, thick fabric. The feminine image was shown as relatively thin and smooth fabric, the sporty images were shown to be moist, flexible and elastic, and the mannish images were relatively rough. The romantic images were shown as relatively thin fabrics. The conclusions inferred from the visual tactile evaluation related to the fabric images recognized by consumers vary by major, so the prior information concerning fabrics and trends can affect the selection of images. The results of this study show that in order to produce clothes suitable for fashion product planning by learning about visual tactility that consumers recognize, fabrics component data displayed in fashion trend books from 2016 to 2018 are needed, so the planner can receive help when selecting the fabrics suitable for each trend.

• 대한전자공학회논문지SP
• /
• 제38권3호
• /
• pp.247-257
• /
• 2001

일반적 칼라 출력 장치는 제한된 범위의 색만을 재현 할 수 있다. 특히 프린터와 같이 색의 재현성이 부족한 장비에서는 하드웨어 특성에 따른 채색 면적이나 인쇄에 사용되는 용지의 특성에 따라 염료의 확산 등에 대해서도 고려해야 한다. 특히, 색을 프린트하기 위해서는 출력 과정에서 색의 선형적인 증가와 감소가 이루어져야 하며 또한 정확한 색의 모델링이 가능하여야 한다. 따라서 본 논문에서는 이를 위해 프린터 출력시 실제 채색 면적과 시각적 색감을 고려할 수 있는 디더링 방법을 제안한다. 제안된 방법은 실제 채색 면적을 고려한 패턴 데이터 베이스를 만들어 이를 디더링 패턴으로 이용하게 된다. 그리고 이 데이터 베이스에서 정확한 패턴을 선택하기 위해서는 인간시각의 색 인지력을 모델링한 대조 민감도 함수를 사용하게 된다. 따라서 본 논문에서는 저해상도의 칼라 출력장치에서도 고화질의 색을 재현할 수 있게 해준다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2012년도 한국컴퓨터종합학술대회논문집 Vol.39 No.1(B)
• /
• pp.324-326
• /
• 2012

기계 학습에서 입력 데이터의 차원을 줄이는 문제(dimension reduction)는 매우 중요한 문제 중의 하나이다. 입력 변수의 차원이 늘어남에 따라 처리해야하는 연산의 수와 계산 복잡도가 급격히 늘어나기 때문이다. 이를 해결하기 위하여 다수의 기계 학습 알고리즘은 명시적으로 차원을 줄이거나(feature selection), 데이터에 약간의 연산을 가하여 차원이 작은 새로운 입력 데이터를 만든다(feature extraction). 반면 사람이 여러 종류의 고차원 센서 데이터를 입력받아 빠른 시간 안에 정확하게 정보를 처리할 수 있는 가장 큰 이유 중 하나는 실시간으로 판단하여 가장 필요한 정보에 집중하기 때문이다. 본 연구는 사람의 정보 처리 과정을 기계 학습 알고리즘에 반영하여, 집중도를 이용하여 효율적으로 데이터를 처리하는 방법을 제시한다. 이 성질을 시각 하이퍼네트워크 모델에 반영하여, 효율적으로 고차원 입력 데이터를 다루는 방법을 제안한다. 실험에서는 시각 하이퍼네트워크를 이용하여 고차원의 이미지 데이터에서 행동을 분류하였다.