• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.106-108
• /
• 2022

이미지 기반 3D 형상 복원에 있어서, 이미지에 보이지 않는 폐색(Occlusion) 영역 부분에 대한 정보가 손실되므로 완전한 복원에 어려움이 있으며, 세밀한 텍스쳐(Texture) 표현이 이루어지지 않고 심한 평활화(Smoothing)나 고립된 노이즈 메쉬(Isolated Noise Mesh) 등 구조적 훼손이 발생한다. 주로 깊은 신경망을 이용하여, 음함수(Implicit Function) 기반 방법은 사전훈련이 완료된 보조 신경망들을 전면부에 배치하거나, Hourglass 등 임베딩(Embedding) 아키텍처를 추가하거나, 또는 표면 법선(Surface Normal)과 같은 환시(Hallucination)를 생성하여 신경망에 입력하기도 한다. 이 논문에서는, 인물의 이미지를 입력받아 색상, 머리카락 및 의상을 포함하는 완전 3D 인간 복원 기술들을 조망해본다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.2
• /
• pp.179-190
• /
• 2001

In this proposed study, observing and analyzing contrast enhanced abdominal CT images, we segmented the liver automatically. We computed the ratio of each gray value from the estimated gray value range. With the average value of mesh image, we distinguished the liver from the noise parts. We divided the region based on immersion simulation. The threshold value is determined from the mesh image which is generated from each gray value portion of the liver and is used in dividing the liver to the noise region. To get the outline of the liver, we generated template image which represents the lump of the liver, and subtracted it from the binary image. With the results we use the proposed algorithm using 8-connectivity instead of the present opening algorithm, to reduce the processing time. We computed the volume from the segmented organ size and presented a clinical demonstration with the animal experiment

• Korean Journal of Remote Sensing
• /
• v.36 no.5_3
• /
• pp.1027-1036
• /
• 2020

In this study, a cloud-based processing method using Agisoft Metashape, a commercial software, and Amazon web service, a cloud computing service, is introduced and evaluated to quickly generate high-precision 3D realistic data from large volume UAV images acquired in disaster sites. Compared with on-premises method using a local computer and cloud services provided by Agisoft and Pix4D, the processes of aerial triangulation, 3D point cloud and DSM generation, mesh and texture generation, ortho-mosaic image production recorded similar time duration. The cloud method required uploading and downloading time for large volume data, but it showed a clear advantage that in situ processing was practically possible. In both the on-premises and cloud methods, there is a difference in processing time depending on the performance of the CPU and GPU, but notso much asin a performance benchmark. However, it wasfound that a laptop computer equipped with a low-performance GPU takes too much time to apply to in situ processing.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.2 s.46
• /
• pp.21-28
• /
• 2007

The facial modeling and animation technology had been studied in computer graphics field. The facial modeling technology is utilized much in virtual reality research purpose of MPEG-4 and so on and movie, advertisement, industry field of game and so on. Therefore, the development of 3D facial model that can do interaction with human is essential to little more realistic interface. We developed realistic and convenient 3D facial modeling system that using a optional facial image only. This system allows easily fitting to optional facial image by using the Korean standard facial model (generic model). So it generates intuitively 3D facial model as controling control points elastically after fitting control points on the generic model wire to the optional facial image. We can confirm and modify the 3D facial model by movement, magnify, reduce and turning. We experimented with 30 facial images of $630{\times}630$ sizes to verify usefulness of system that developed.

• Journal of Broadcast Engineering
• /
• v.17 no.5
• /
• pp.838-850
• /
• 2012

In this paper, we propose a novel algorithm that generates multi-view images by using various image features obtained from the given stereoscopic images. In the proposed algorithm, we first create an intensity gradient saliency map from the given stereo images. And then we calculate a block-based optical flow that represents the relative movement(disparity) of each block with certain size between left and right images. And we also obtain the disparities of feature points that are extracted by SIFT(scale-invariant We then create a disparity saliency map by combining these extracted disparity features. Disparity saliency map is refined through the occlusion detection and removal of false disparities. Thirdly, we extract straight line segments in order to minimize the distortion of straight lines during the image warping. Finally, we generate multi-view images by grid mesh-based image warping algorithm. Extracted image features are used as constraints during grid mesh-based image warping. The experimental results show that the proposed algorithm performs better than the conventional DIBR algorithm in terms of visual quality.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2008.11a
• /
• pp.117-120
• /
• 2008

본 논문은 스테레오 비디오 데이터 압축 및 복원을 위한 그리드(Grid) 기반 2D 워핑 방법을 제안한다. 스테레오 비디오에 대한 데이터 표현 방법으로 깊이지도 및 매쉬(mesh)를 이용한 방법이 주로 사용되어 왔으며 매쉬를 이용한 방법은 두 영상간의 매칭되는 노드를 이용하여 데이터 압축 효율을 높일 수 있다. 그러나, 두 영상에서 매칭되는 노드의 위치를 찾는 것은 매우 어려운 일일 뿐만 아니라 매쉬에 의해서 워핑된 영상과 목적이 되는 스테레오 영상의 좌측 또는 우측 영상간의 왜곡이 불가피하다. 따라서 이러한 왜곡을 보정하기 위하여 잔여영상(Residual image) 정보를 추가로 요구하게 된다.제안된 논문은 이러한 잔여영상 정보를 최소화 하기 위하여 반복적으로 2D워핑을 수행하며 최적화된 워핑 영상을 생성함으로써 목적영상과의 오차를 최소로 유지하여 추가정보인 잔여 영상의 데이터 용량을 최소화 한다. 전체영상에 대하여 2D워핑을 수행하며 각각의 노드를 변경하는 것은 많은 비용을 감수해야 하기 때문에 오차영역에 대하여 지역단위로 분할하고 단계적으로 최적화를 이루는 분할정복 방법을 사용하였다. 본 논문의 실험에서는 스테레오 영상에 대하여 각각의 신호대 잡음비(PSNR)를 통해 제안한 방법의 품질을 평가하였을 뿐만 아니라 기존의 메쉬 기반한 방법과 깊이지도를 이용한 방법과의 데이터량을 비교하였다. 실험결과를 통하여 제안한 방법의 데이터 압축의 효율성 및 품질의 우수성을 확인하였다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.1_2
• /
• pp.20-32
• /
• 2001

각 장기는 고유한 명암값의 범위와 각 명암값에 대한 서로 다른 비율을 지니고 있으므로 제안된 연구에서는 이러한 명암값의 비율을 이용하여 장기의 영역과 노이즈를 구분할 수 있도록 하였다. 장기의 영역을 세 종류의 메쉬영상으로 표현하여 이들의 유니온 영상으로 장기의 전반적인 형태인 템플리트를 생성하였다. 템플리트 방식은 기존의 방식에서 명암값의 범위가 같은 노이즈의 제거가 어려운 단점을 해결하여 장기의 영역만을 분리할 수 있었다. 장기의 위치를 탐색하기 위한 위치탐색과정에서는 장기의 존재여부의 파악과 함께 분리된 장기까지 추적할 수 있도록 하였다. 외곽선 표현을 위해서는 템플리트로 이진영상에서 서브트랙션(subtraction)하는 방법을 사용하여 장기의 말단부위까지 세밀하게 표현하였다. 제안된 연구에서 사용된 오프닝과 클로징 방법으로 기존의 structuring element를 사용하는 방법에 비해 처리속도를 단축시킬 수 있었다. 추출된 장기의 면적을 토대로 체적계산을 시행하였고 동물실험을 통하여 임상 실험치를 제시하였다.

• Journal of the Korea Computer Graphics Society
• /
• v.15 no.3
• /
• pp.27-33
• /
• 2009

In terrain visualization, the quadtree is the most frequently used data structure for progressive mesh generation. The quadtree provides an efficient level-of-detail selection and view frustum culling. However, most applications using quadtrees are performed by the CPU, since the hierarchical data structure cannot be manipulated in a programmable rendering pipeline. For this reason, quadtree-based methods show lower performance and higher dependancy of CPU in comparison to GPU-based methods. We present a quadtree-based terrain-rendering method for GPU execution that uses vertex multiplication. It offers higher performance than previous CPU-based quadtree methods, without loss of image quality.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.40 no.6
• /
• pp.384-392
• /
• 2003

The algorithm for calculating the base-collector breakdown voltage of NPN BJT for integrated circuits is proposed. The method of three-dimensional mesh generation to minimize the time required for device simulation is presented and the method for calculating the breakdown voltage using solutions of the Poisson´s equation is presented. To verify the proposed method, the breakdown voltage between base and collector of NPN BJT using 20V process and 30V process is compared with the measured data. The breakdown voltage from the proposed method of NPN BJT using 20V process shows an averaged relative error of 8.0% compared with the measured data and the breakdown voltage of NPN BJT using 30V process shows an averaged relative error of 4.3% compared with the measured data.

• Journal of Korean Association for Spatial Structures
• /
• v.12 no.1
• /
• pp.43-50
• /
• 2012

Recently, due to the advanced computer technology, momental architectures have been designed and built using features that are very sophisticated. People's interest in free-form structural system has increased steadily not only nationwide, but also worldwide. However, there were many difficulties in the materialization of free-form structural system owing to the lack of technique and research. To solve this problem, this study performs the interface between the 3D modeling program and the optimization program. In the 3D modeling program, it is possible to automatic mesh generation and immediately to information extraction. It performs the shape optimization. Consequently, this research designs the example model and performs optimization in order to verify the developed interface module.