• Journal of Broadcast Engineering
• /
• v.21 no.6
• /
• pp.878-888
• /
• 2016

This paper proposes an efficient method to remove the boundary artifacts of rendered views caused by damaged depth maps in the 3D video system. First, characteristics of boundary artifacts with the compression noise in depth maps are carefully studied. Then, the artifacts suppression method is proposed by the iterative projection onto convex sets (POCS) algorithm with setting the convex set in pixel and frequency domain. The proposed method is applied to both texture and depth maps separately during view rendering. The simulation results show the boundary artifacts are greatly reduced with improving the quality of synthesized views.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.130-133
• /
• 2010

멀티미디어 신호 처리 분야의 급속한 발전으로 인해 3차원 TV(3DTV)는 차세대 방송 시스템 시장에서 가장 주목을 받는 제품이 되었다. 3DTV는 사용자에게 입체감을 제공하여 사용자가 마치 그곳에 있는 듯한 효과를 줄 수 있다. 이에 최근 다시점 컬러 영상과 그에 대응되는 깊이 영상을 이용하여 다양한 위치에서의 입체 영상을 제공하는 기술이 활발히 진행되고 있다. 본 논문에서는 주어지는 다시점 컬러 영상과 깊이 영상의 정보를 동시에 이용하여 다시점 영상 부호화의 효율을 높이는데 그 목적이 있다. 이를 위해 시점 생성 기법을 통해 다시점 영상의 부호화 시 참조하는 주변 시점의 영상들로부터 계산되는 변이 벡터 정보를 줄이는 방법을 제안한다. 실험 결과, 기존의 다시점 비디오 부호화에 비해서 약 0.5~1.5 dB정도의 화질 개선이 이뤄졌고, 평균 17.5%의 비트율 감소를 보였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.369-372
• /
• 2011

본 논문에서는 가상 시점 생성 시 발생하는 홀 영역의 메움 기법에 관하여 기하학적 제약조건을 이용한 예제 기반 영상 인페인팅 기술을 제안한다. 생성된 가상 시점이 실제 카메라들의 중간에 위치하는 경우, 이 때 발생하는 홀 영역은 시점 간 보상으로 메울 수 있다. 하지만, 가상 시점이 실제 카메라들의 Field-of-View를 벗어난 경우에는 홀 영역에 대한 정보가 없기 때문에 이를 메우는데 어려운 문제가 따른다. 이러한 문제를 해결하기 위해 제안한 방법은 예제 기반의 영상 인페인팅 기술을 이용하여 홀 영역을 메운다. 제안한 방법은 홀 영역의 기하학적 특성을 제약조건으로 우선순위(Priority)를 계산한다. 최적의 패치를 찾기 위해 컬러 정보뿐만 아니라 깊이 정보(Depth)를 활용한다. 또한, 전경 물체의 텍스쳐로 인하여 잘못된 패치가 선택되는 문제를 해결하기 위해 적응적으로 패치 크기를 변경한다. 실험 결과에서 제안한 방법은 일반적인 홀 메움 방식과 비교하여 객관적 평가와 주관적 평가 모두에서 시각적으로 우수한 결과를 보여주었다.

• Electronics and Telecommunications Trends
• /
• v.35 no.5
• /
• pp.112-122
• /
• 2020

In 3D computer graphics, a depth map is an image that provides information related to the distance from the viewpoint to the subject's surface. Stereo sensors, depth cameras, and imaging systems using an active illumination system and a time-resolved detector can perform accurate depth measurements with their own light sources. The 3D image information obtained through the depth map is useful in 3D modeling, autonomous vehicle navigation, object recognition and remote gesture detection, resolution-enhanced medical images, aviation and defense technology, and robotics. In addition, the depth map information is important data used for extracting and restoring multi-view images, and extracting phase information required for digital hologram synthesis. This study is oriented toward a recent research trend in deep learning-based 3D data analysis methods and depth map information extraction technology using a convolutional neural network. Further, the study focuses on 3D image processing technology related to digital hologram and multi-view image extraction/reconstruction, which are becoming more popular as the computing power of hardware rapidly increases.

• KIEAE Journal
• /
• v.13 no.1
• /
• pp.109-123
• /
• 2013

The study is to probe a technical alternative to enhance the reliability and accuracy of the results of various landscape simulations. This study to present technical criteria that are necessary in each stage of target site analysis, picture taking, and computer synthesis and, through these to present supplementary plans to enhance the reliability and accuracy of landscape simulations. In order to derive more practical and empirical results in terms of the reliability of the results of landscape simulations, examples that actually passed landscape review were selected. With regard to study process, an analysis was made first to analyze the landscape report data of designs that passed the review to analyze their characteristics, to be followed by an integrated analysis of problems that were revealed in various landscape simulations. Important factors that affect directly the work of landscape simulations such as the specification of camera lenses that were used in picture taking, distance, and angle. Design the work was carried out using Auto CAD, 3DS Max, and Photoshop program in the same way as in actual design. For verification of their accuracy and reliability, the results were entrusted to experts who have implemented similar jobs. To seduce differences from those too landscape simulations that conduct trial experiment of the virtual space that are to be created in the future with accurate numerical values.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.12C
• /
• pp.738-743
• /
• 2011

For the high quality 3-D broadcasting, depth maps are important data. Although commercially available depth cameras capture high-accuracy depth maps in real time, their resolutions are much smaller than those of the corresponding color images due to technical limitations. In this paper, we propose the depth map up-sampling method using a high-resolution color image and a low-resolution depth map. We define a random walk probability model in an operation unit which has nearest seed pixels. The proposed method is appropriate to match boundaries between the color image and the depth map. Experimental results show that our method enhances the depth map resolution successfully.

• Korean Journal of Optics and Photonics
• /
• v.20 no.6
• /
• pp.319-327
• /
• 2009

We introduce a new design technique by treating a two-group zoom lens system with a wide angle of view. First, the concept of the global optimization is introduced in the initial design stage, and from this, the global design technique is completed by analyzing and summarizing large quantities of modern design data. That is, we define the global structure function to achieve a new conceptual design technique for global optimization. And the function is put in a simple form by referring lots of patent data, manipulated with other algebraic equations, and solved finally such that we obtain the global solution region. The global solution region corresponds to the global optimization and suggests insightful systematized directions for the design of two-group zoom lens systems. These directions are attractive compared to global optimization.

• Journal of Information Display
• /
• v.6 no.4
• /
• pp.26-34
• /
• 2005

In this paper, a new rectification scheme to transform the uncalibrated stereo image pair into the calibrated one is suggested and its performance is analyzed by applying this scheme to the reconstruction of the intermediate views for multi-view stereoscopic display. In the proposed method, feature points are extracted from the stereo image pair by detecting the comers and similarities between each pixel of the stereo image pair. These detected feature points, are then used to extract moving vectors between the stereo image pair and the epipolar line. Finally, the input stereo image pair is rectified by matching the extracted epipolar line between the stereo image pair in the horizontal direction. Based on some experiments done on the synthesis of the intermediate views by using the calibrated stereo image pairs through the proposed rectification algorithm and the uncalibrated ones for three kinds of stereo image pairs; 'Man', 'Face' and 'Car', it is found that PSNRs of the intermediate views reconstructed from the calibrated images improved by about 2.5${\sim}$3.26 dB than those of the uncalibrated ones.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.144-145
• /
• 2020

본 논문에서는 정밀한 3차원 복원 및 시점 합성을 위해 매칭 비용을 반복적으로 업데이트하는 Generalized Soft 3D Reconstruction (GenSoft3D) 알고리즘을 제안한다. 먼저 다시점 영상들과 카메라 자세정보가 주어지면 GenSoft3D는 볼륨 기반의 다시점 스테레오 매칭 알고리즘으로 시점별 초기 매칭 비용 볼륨 및 시차 맵을 계산한다. 그 후 정제 과정에서 각 시점은 모든 시차 맵을 이용하여 표면 확률 및 가시 확률을 계산한다. 표면 확률은 초기 매칭 비용 업데이트에 사용하며, 가시 확률은 폐색 영역의 정확한 시차를 계산하기 위해 사용된다. 해당 정제 과정을 일정 횟수 반복할 경우 시점별 고정밀의 시차 맵 획득이 가능하다. 또한 시차 맵의 정확도가 향상됨에 따라 정확한 시점 합성이 가능하다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.5
• /
• pp.77-87
• /
• 2018

Traffic Density is the most important of the three primary macroscopic traffic stream parameters, because it is most directly related to traffic demand(Traffic Engineering, 2004). It is defined as the number of existing vehicles within a given distance at a certain time. However, due to weather, road conditions, and cost issues, collecting density directly on the field is difficult. This makes studies of density less actively than those of traffic volume or velocity. For these reasons, there is insufficient attempts on divers collecting methods or researches on the accuracy of measured values. In this paper, we used the 'Density Measuring System' based on the synthesise technology of several camera images as a method to measure density. The collected density value by the 'Density Mesuring System' is selected as the true value based on the density define, and this value was compared with the density calculated by the traditional measurement methods. As a result of the comparison, the density value using the fundamental equation method is the closest to the true value as RMSE shows 1.8 to 2.5. In addition, we investigated some issues that can be overlooked easily such as the collecting interval to be considered on collecting density directly by calculating the moment density and the average density. Despite the actual traffic situation of the experiment site is LOS B, it is difficult to judge the real traffic situation because the moment density values per second are observed max 16.0 (veh/km) to min 2.0 (veh/km). However, the average density measured for 15 minutes at 30-second intervals was 8.3-7.9 (veh/km) and it indicates precisely LOS B.