• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.192-195
• /
• 2008

본 논문에서는 발전하는 3D 그래픽스 기술을 이용하여 문화재의 도면 실루엣을 생성하는 방법을 제안하고자 한다. 3D 스캐너로 정밀 실측된 3D 데이터를 이용하여 문화재의 도면을 생성하기 위한 벡터 실루엣(Silhouette) 추출 과정은 다음과 같다. 먼저 실측된 3D 데이터를 정규화 된 3D공간으로 이동하고, 이동 후에는 데이터에 존재하는 모든 에지(edge)를 검출하여 에지리스트(edge list)를 생성한다. 생성된 에지리스트는 다시 윤곽에지(Contour edge)와 주름에지(Crease edge)로 분류하는데, 윤곽에지는 문화재의 윤곽 실루엣을 형성하는데 이용하고, 윤곽에지를 제외한 주름에지는 문화재의 표면 특징을 나타내는 내부문양 실루엣을 형성하는데 이용한다. 내부문양 실루엣은 사용자가 입력하는 임계값과 주름에지를 구성하는 두면의 방향 벡터의 내적을 비교하여 추출한다. 추출한 벡터 실루엣은 윤곽 실루엣과 내부문양 실루엣으로 구분되며, 두 벡터 실루엣을 이용함으로써 문화재의 구조적 해석과 표면의 특징을 해석할 수 있는 도면 실루엣 생성이 가능했다.

• The Journal of the Korea Contents Association
• /
• v.8 no.12
• /
• pp.10-19
• /
• 2008

In this paper, we propose an automated silhouette extraction method that can effectively extract inner-patterns and silhouettes from 3D scan data of cultural asset. First of all, after creating the edge list of 3D vector data, we decide contour edge and crease edge according to viewpoint. In the next step, after extracting surface silhouette by investigating the vector variation of adjacent faces in crease edge, we finally extract the contour silhouette and surface silhouette for generating the blueprint of cultural asset. To evaluate the performance of the proposed silhouette extraction method, we performed experiments of silhouette extraction using a traditional tile model, a car model and a stone monument model. Comparing with the conventional threshold-based silhouette extraction method, the proposed method extracted more distinct and clear surface silhouettes and inner-patterns by effectively removing meaningless edges, such as noise.

• Cartoon and Animation Studies
• /
• s.32
• /
• pp.1-21
• /
• 2013

Silhouette Animation has been recognized as a genre of animation since the very beginning of the animation history, and also Its segmental movement and Aesthetic expression has led a variety of semantic interpretations. Especially the French animation director, Michel Ocelot, recently integrated 3-dimensional digital to the silhouettes animation, and it extended the possibility of the silhouettes animation in many aspects. In his latest animation feature, , he showed how he made changes in 3-dimension by creating and evolving his own way and style of silhouette animation. Although mainstream digital animations preferably to show realistic images and motion, Michel Ocelot used very selective movement, subjective digital colors and extended space which couldn't be expressed in the way of creating traditional style of silhouette animation. This alternative slow movement and the unique aesthetics in 3-dimension emphasize the unconscious elements of color, composition, patterns, and it provides digitally enhanced images and pictorial impression. In addition, the acquisition of digital three-dimensional use of space made possible to provides the wider formative imagination to the audience. In this paper, we analyzed aesthetic characteristics of the digital silhouette animation, (2011), specially focusing on the aspects of Movement, Image, Space, which could not be found in the traditional silhouette animation. It is significant to obtain diversity of the future digital animation and its positive development. In addition, this provides opportunity to explore Michel Ocelot's new experiments and animation philosophy.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.923-926
• /
• 2011

본 연구에서는 움직이는 물체가 있는 비디오에서 검출된 전경 영상(실루엣)을 토대로 사람을 추적하고 추적된 사람의 실루엣 형상을 통하여 활동성을 인식하는 실시간 감시 시스템에 적용 가능한 사람의 행동을 인식하고 분석하고자 한다. 전경에서 블랍(사람)을 검출하는 방법은 기존에 연구했던 차영상을 이용하였고, 검출된 블랍을 대상으로 사람임을 판단하고 사람인 경우 검출된 블랍의 실루엣을 이용한 기존의 자세 추정 기법에 추가적으로 4가지 특징들을 추가하여 사람의 행동을 분석한다. 각 파라미터들은 임계치를 통하여 구분하였다. 본 논문에서는 사람의 행동은 크게 네 가지의 경우로 {Standing, Bending/Crawling, Laying down, Sitting} 분류한다. 제안된 특징 파라미터들을 추가한 방법은 기존의 실루엣 기반의 자세 추정 기법만을 사용하는 것보다 좀더 높은 인식율을 보여주었다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.6C
• /
• pp.593-603
• /
• 2009

In this paper, we propose a new integrated computer vision system designed to track multiple human beings and extract their silhouette with an active stereo camera. The proposed system consists of three modules: detection, tracking and silhouette extraction. Detection was performed by camera ego-motion compensation and disparity segmentation. For tracking, we present an efficient mean shift based tracking method in which the tracking objects are characterized as disparity weighted color histograms. The silhouette was obtained by two-step segmentation. A trimap is estimated in advance and then this was effectively incorporated into the graph cut framework for fine segmentation. The proposed system was evaluated with respect to ground truth data and it was shown to detect and track multiple people very well and also produce high quality silhouettes. The proposed system can assist in gesture and gait recognition in field of Human-Robot Interaction (HRI).

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.13-22
• /
• 2023

In this paper, we present a method for estimating full 3D shapes from given 2D silhouette images of human bodies. Because the silhouette only consists of the partial information on the true shape, it is an ill-posed problem. To address the problem, we use the statistical human shape space obtained from the existing large 3D human shape database. The method consists of three steps. First, we extract the boundary pixels and their appropriate normal vectors from the input silhouette images. Then, we initialize the correspondences of each pixel to the vertex of the statistically-deformable 3D human model. Finally, we numerically optimize the parameters of the statistical model to fit best to the given silhouettes. The viability and the robustness of the method is demonstrated with various experiments.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.505-507
• /
• 2001

실루엣 에지를 찾거나 디스플레이하는 것은 컴퓨터 비전에서부터 비사실적인 렌더링(nonphotorealistic rendering)에 이르기까지 여러 응용분야에서 중요한 의미를 가진다. NPR(nonphotorealistic rendering)에서 가장 초점을 두는 것은 컴퓨터로 렌더링된 결과가 사람이 그린듯한 효과를 줄 수 있느냐에 있다. 본 논문에서는 점진적 메쉬와 실루엣 에지 렌더링을 통해 사람이 임의의 객체를 점진적으로 스케치하는 효과를 보여 줄 수 있는 방법을 제시한다. 다단계 메쉬 생성은 대용량의 메쉬를 효율적으로 다루기 위한 것으로 점진적인 렌더링이나 전송, 상세도, 제어 등에 이용되는 기술이다. 메쉬 간략화에서는 볼륨과 정계를 보존 할 수 있는 간략화 기법을 이용하였다. 간략화 수행시 원본 메쉬에 대한 히스토리를 저장할 수 있는 점진적 메쉬(Progressive Mesh)구조를 구성한 뒤 기본메쉬를 점진적으로 상세하게 해 나감으로써 실루엣 렌더링을 수행하게 된다. 실루엣 에지에 대해 파라미터화된 브러쉬 함수(Parameterized Brush Functions)를 적용하여 다양한 스타일로 점진적인 스케칭 효과를 낼 수 있는 기법을 제시한다.

• Journal of Digital Contents Society
• /
• v.10 no.3
• /
• pp.381-388
• /
• 2009

This paper proposes a denoising method for the contour edges and crease edges of silhouette obtained from 3D scanned data of stone cultural heritages. It is often the case that the silhouette involves noise in the form of short-length line segments, due to rough surfaces of stone cultural heritages, weathering, and technical difficulties arising in data acquisition. Thus the removal of the short-length line segments from the contour edges and crease edges can result in a clear and accurate silhouette. An efficient computational algorithm is introduced to count the continuity of line segments; edges having not more than 3 line segments are removed. It has been verified that the new method is more effective than threshold-based silhouette extraction methods for stone heritages. Our method is applicable for various other data which are deteriorated by short-length line segments.

• The Journal of the Korea Contents Association
• /
• v.12 no.12
• /
• pp.577-586
• /
• 2012

This research was focused on jean jacket and jean pant design characteristics in the collection. To offer a basic proposal for the development of jean jackets and pants, pictures of fashion web pages from 2007 to 2011 were used, and data were analysed by the usage of the frequency and percentage of the SPAW Statistics 18. The results of the research were as follows. Men's jackets were mainly medium in length with a tetragonal silhouette and simple detail. On the contrary, women's jackets were mainly of an X silhouette, short in length, and with varied details. Men's jean pants were mainly represented by a straight, comfortable silhouette while women's jean pants were characterized by a variety of silhouettes, fit, and lengths.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.852-857
• /
• 2005

This paper discusses a human body extraction method for intelligent robot system. The intelligent robot system requires more robust silhouette extraction method because it has internal vibration and low resolution. The new hybrid silhouette extraction method is proposed to overcome this constrained environment. The temporal and gradient information is combined as hybrid silhouette. The motion region model is used to adjust combining parameters in hybrid silhouette. Finally, the experimental results show the superiority of the proposed method.