• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.193-195
• /
• 2022

본 논문에서는 3차원 스캔 없이 이미지 입력만을 사용한 개인 체형을 고려한 모바일 가상 착용 시스템의 전체 과정을 설계하고 개발하였다. 이를 위하여 이미지상 인물의 자세와 체형의 추정을 통하여 3차원 인체모델(SMPL)을 추정하는 최근의 방식을 이용하였고, 앞 뒷면 의상 이미지를 2차원 texture 매핑과 평면 triangle mesh로 복원하고 의상 봉제 (sewing) 시뮬레이션을 사용하여 3차원 의상 모델을 생성하는 방법을 새롭게 개발하였다. 또한 이를 활용한 3차원 개인화된 가상 착용 모바일 앱과 서비스를 Flask와 iOS 환경에서 SceneKit을 활용하여 개발하였다. 이를 통하여 단순히 의상의 매칭과 스타일 뿐 아니라 사이즈에 따른 착용 Fit을 구매 전에 확인할 수 있는 전체 서비스를 실현 및 검증하였다.

• Molecules and Cells
• /
• v.38 no.11
• /
• pp.975-981
• /
• 2015

Precise 3D spatial mapping of cells and their connections within living tissues is required to fully understand developmental processes and neural activities. Zebrafish embryos are relatively small and optically transparent, making them the vertebrate model of choice for live in vivo imaging. However, embryonic brains cannot be imaged in their entirety by confocal or two-photon microscopy due to limitations in optical range and scanning speed. Here, we use light-sheet fluorescence microscopy to overcome these limitations and image the entire head of live transgenic zebrafish embryos. We simultaneously imaged cranial neurons and blood vessels during embryogenesis, generating comprehensive 3D maps that provide insight into the coordinated morphogenesis of the nervous system and vasculature during early development. In addition, blood cells circulating through the entire head, vagal and cardiac vasculature were also visualized at high resolution in a 3D movie. These data provide the foundation for the construction of a complete 4D atlas of zebrafish embryogenesis and neural activity.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.37
• /
• pp.44.1-44.6
• /
• 2015

Background: The present study introduces the design and fabrication of a simple surgical guide with which to perform genioplasty. Methods: A three-dimensional reconstruction of the patient's cranio-maxilla region was built, with a dentofacial skeletal model, then derived from CT DICOM data. A surgical simulation was performed on the maxilla and mandible, using three-dimensional cephalometry. We then simulated a full genioplasty, in silico, using the three-dimensional (3D) model of the mandible, according to the final surgical treatment plan. The simulation allowed us to design a surgical guide for genioplasty, which was then computer-rendered and 3D-printed. The manufactured surgical device was ultimately used in an actual genioplasty to guide the osteotomy and to move the cut bone segment to the intended location. Results: We successfully performed the osteotomy, as planned during a genioplasty, using the computer-aided design and computer-aided manufacturing (CAD/CAM) surgical guide that we initially designed and tested using simulated surgery. Conclusions: The surgical guide that we developed proved to be a simple and practical tool with which to assist the surgeon in accurately cutting and removing bone segments, during a genioplasty surgery, as preoperatively planned during 3D surgical simulations.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1321-1330
• /
• 2023

This study attempts to address the problem of 3D pose estimation for multiple human objects through a single image generated during the character development process that can be used in augmented reality. In the existing top-down method, all objects in the image are first detected, and then each is reconstructed independently. The problem is that inconsistent results may occur due to overlap or depth order mismatch between the reconstructed objects. The goal of this study is to solve these problems and develop a single network that provides consistent 3D reconstruction of all humans in a scene. Integrating a human body model based on the SMPL parametric system into a top-down framework became an important choice. Through this, two types of collision loss based on distance field and loss that considers depth order were introduced. The first loss prevents overlap between reconstructed people, and the second loss adjusts the depth ordering of people to render occlusion inference and annotated instance segmentation consistently. This method allows depth information to be provided to the network without explicit 3D annotation of the image. Experimental results show that this study's methodology performs better than existing methods on standard 3D pose benchmarks, and the proposed losses enable more consistent reconstruction from natural images.

• Smart Media Journal
• /
• v.3 no.4
• /
• pp.22-27
• /
• 2014

Three-dimensional image reconstruction allows us to represent real objects in the virtual space and observe the objects at arbitrary view points. This technique can be used in various application areas such as education, culture, and art. In this paper, we propose an implementation method of the high-quality three-dimensional object using multiple Kinect cameras released from Microsoft. First, We acquire color and depth images from triple Kinect cameras; Kinect cameras are placed in front of the object as a convergence form. Because original depth image includes some areas where have no depth values, we employ joint bilateral filter to refine these areas. In addition to the depth image problem, there is an color mismatch problem in color images of multiview system. In order to solve it, we exploit an color correction method using three-dimensional geometry. Through the experimental results, we found that three-dimensional object which is used the proposed method is more naturally represented than the original three-dimensional object in terms of the color and shape.

• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1447-1455
• /
• 2017

In a trawl-net simulation, it is very important to process the physical phenomenons resulting from real collisions between a net and fishes. However, because it is very difficult to reconstruct the surface with mass points, many researchers have generally detect the collision using an approximation model employing a sphere, a cube or a cylinder. These approaches occur often result in inaccurate movements of a fish due to the difference between a real-net and a designed-net. So, many systems have manually adjusted a net surface based on actual measurements of mass points. These methods are very inefficient because it needs much times in an adjustment and also causes more incorrect inputs according to a rapid increment in the number of points. Therefore, in this paper, we propose a reconstruction method that it semi-automatically reconstructed trawl-net surfaces using the equation of motion at each mass point in a mass-spring model. To get an easy start in a beginning step of the spread, it enables users to get interactive adjustment on each mass point. We had designed a trawl-net model using geometrical structures of trawl-net and then automatically reconstructed the trawl-net surface using scale-space meshing techniques. Last, we improve the accuracy of reconstructed result by correction user interaction.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.43 no.4
• /
• pp.256-261
• /
• 2017

Objectives: The concept of natural head position (NHP) was first introduced by Broca in 1862, and was described as a person's stable physiologic position "when a man is standing and his visual axis is horizontal." NHP has been used routinely for clinical examination; however, a patient's head position is random during cone-beam computed tomography (CBCT) acquisition. To solve this problem, we developed an accelerometer to record patients' NHP and reproduce them for CBCT images. In this study, we also tested the accuracy and reproducibility of our accelerometer. Materials and Methods: A total of 15 subjects participated in this study. We invented an accelerometer that measured acceleration on three axes and that could record roll and pitch calculations. Recorded roll and pitch data for each NHP were applied to a reoriented virtual image using three-dimensional (3D) imaging software. The data between the 3D models and the clinical photos were statistically analyzed side by side. Paired t-tests were used to statistically analyze the measurements. Results: The average difference in the angles between the clinical photograph and the 3D model was $0.04^{\circ}$ for roll and $0.29^{\circ}$ for pitch. The paired t-tests for the roll data (P=0.781) and the pitch data (P=0.169) showed no significant difference between the clinical photographs and the 3D model (P>0.05). Conclusion: By overcoming the limitations of previous NHP-recording techniques, our new method can accurately record patient NHP in a time-efficient manner. Our method can also accurately transfer the NHP to a 3D virtual model.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.21 no.3
• /
• pp.229-236
• /
• 2003

This study recorded an object using a digital video recorder, and then tried to estimate 3-D positional information and to reconstruct an image. Firstly, the accuracy of measurement results from a video recorder was evaluated and tested for an applicability, then it applied to a real object to construct 3-D digital model. This study assumed that there is no lens distortion in a video recorder, and all bundles should precisely pass through the projection center of a lens. The image size for orientations is determined by the size of CCD chip and the number of pixels. The average squared error from the result by a digital video recorder and that by triangular survey from 1-second theodolite shows 0.0173m error in x,y coordinates. Without knowing the accurate information on the lens distortion and the coordinates of the projection center, this study reasonably produces acceptable results in the reconstruction of 3-D model. In consequence, this study found that the image from a digital video camera can be reconstructed 3-D model only from the information on a camera type.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.16 no.1
• /
• pp.12-16
• /
• 2012

Purpose : In conventional PET image reconstruction, iterative reconstruction methods such as OSEM (Ordered Subsets Expectation Maximization) have now generally replaced traditional analytic methods such as filtered back-projection. This includes improvements in components of the system model geometry, fully 3D scatter and low noise randoms estimates. SharpIR algorithm is to improve PET image contrast to noise by incorporating information about the PET detector response into the 3D iterative reconstruction algorithm. The aim of this study is evaluation of SharpIR reconstruction method in PET/CT. Materials and Methods: For the measurement of detector response for the spatial resolution, a capillary tube was filled with FDG and scanned at varying distances from the iso-center (5, 10, 15, 20 cm). To measure image quality for contrast recovery, the NEMA IEC body phantom (Data Spectrum Corporation, Hillsborough, NC) with diameters of 1, 13, 17 and 22 for simulating hot and 28 and 37 mm for simulating cold lesions. A solution of 5.4 kBq/mL of $^{18}F$-FDG in water was used as a radioactive background obtaining a lesion of background ratio of 4.0. Images were reconstructed with VUE point HD and VUE point HD using SharpIR reconstruction algorithm. For the clinical evaluation, a whole body FDG scan acquired and to demonstrate contrast recovery, ROIs were drawn on a metabolic hot spot and also on a uniform region of the liver. Images were reconstructed with function of varying iteration number (1~10). Results: The result of increases axial distance from iso-center, full width at half maximum (FWHM) is also increasing in VUE point HD reconstruction image. Even showed an increasing distances constant FWHM. VUE point HD with SharpIR than VUE point HD showed improves contrast recovery in phantom and clinical study. Conclusion: By incorporating more information about the detector system response, the SharpIR algorithm improves the accuracy of underlying model used in VUE point HD. SharpIR algorithm improve spatial resolution for a line source in air, and improves contrast recovery at equivalent noise levels in phantoms and clinical studies. Therefore, SharpIR algorithm can be applied as through a longitudinal study will be useful in clinical.

• Journal of the Korea Computer Graphics Society
• /
• v.15 no.2
• /
• pp.1-8
• /
• 2009

This paper introduces a virtual dress up system, according to user's input model and garment image. At first step, we deform the garment image by using skeleton structures and ARAP method. Next step, sampling the boundary points and find their matching vertices which are used for optimizing the boundary fitting. In 2D rendering of the dress up, they have some unrealistic results, so we reconstruct the garment mesh to the 3D mesh. Rendering from the reconstructed 3D mesh, we can get the final dress up result. We present that our system produce a visually plausible and well-fitted virtual dress up results.