• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.142-146
• /
• 2023

We present a method for generating 3D structures and rendering objects by combining VAE (Variational Autoencoder) and GAN (Generative Adversarial Network). This approach focuses on generating and rendering 3D models with improved quality using residual learning as the learning method for the encoder. We deep stack the encoder layers to accurately reflect the features of the image and apply residual blocks to solve the problems of deep layers to improve the encoder performance. This solves the problems of gradient vanishing and exploding, which are problems when constructing a deep neural network, and creates a 3D model of improved quality. To accurately extract image features, we construct deep layers of the encoder model and apply the residual function to learning to model with more detailed information. The generated model has more detailed voxels for more accurate representation, is rendered by adding materials and lighting, and is finally converted into a mesh model. 3D models have excellent visual quality and accuracy, making them useful in various fields such as virtual reality, game development, and metaverse.

• Journal of Biomedical Engineering Research
• /
• v.21 no.5
• /
• pp.519-526
• /
• 2000

볼륨 렌더링은 3D 의료영상 데이터를 가시화하는 중용한 기법 중 하나이다. 그러나 볼륨 렌더링을 실시간으로 이룰 때, 많은 계산량을 필요로하는 것이 볼률 렌더링을 사용하는데 걸림돌이 되고 있다. 이 논문에서는 Superscalar와 VLIM(Very Long Instruction Word)의 구조를 가지고 있어 동시에 8개의 명령어 수행이 가능한 TI사의 TMS320C6201 DSP를 이용하여 3D 초음파 영상의 쉬어-웝 볼륨 렌더링을 구현하였다. 쉬어-웝 방법을 DSP 상에서 최적으로 구현하기 위하여 ray map 방법, one-to-four ray casting, ？디 skipping 방법을 제안하였다. 제안한 방법들을 이용한 볼륨 렌더링과 적용하지 않은 기존의 알고리즘을 DSP에 구현하여 PSNR과 렌더링 시간의 비교·평가를 통해 만족할 만한 영상 화질에 빠른 렌더링 성능을 얻을 수 있음을 보여주었다.

• Journal of Korea Game Society
• /
• v.20 no.3
• /
• pp.75-84
• /
• 2020

This paper introduces an interactive indirect illumination algorithm which considers indirect visibility. First, a small number of rays are emitted on hemisphere of the current pixel to obtain the first intersection. If this point is directly illuminated by the light source, its illuminated color is collected. Second, in order to approximate the indirect visibility, a 3D ray marching algorithm, which is based on a hierarchy structure, is used to accelerate the ray-voxel intersection. Third, the indirect images are denoised by an edge-avoiding filtering with a local means replacement method.

• Journal of Ocean Engineering and Technology
• /
• v.35 no.3
• /
• pp.183-190
• /
• 2021

Tomographic particle image velocimetry (PIV) is a widely used method that measures a three-dimensional (3D) flow field by reconstructing camera images into voxel images. In 3D measurements, the setting and calibration of the camera's mapping function significantly impact the obtained results. In this study, a camera self-calibration technique is applied to tomographic PIV to reduce the occurrence of errors arising from such functions. The measured 3D particles are superimposed on the image to create a disparity map. Camera self-calibration is performed by reflecting the error of the disparity map to the center value of the particles. Vortex ring synthetic images are generated and the developed algorithm is applied. The optimal result is obtained by applying self-calibration once when the center error is less than 1 pixel and by applying self-calibration 2-3 times when it was more than 1 pixel; the maximum recovery ratio is 96%. Further self-correlation did not improve the results. The algorithm is evaluated by performing an actual rotational flow experiment, and the optimal result was obtained when self-calibration was applied once, as shown in the virtual image result. Therefore, the developed algorithm is expected to be utilized for the performance improvement of 3D flow measurements.

• Current Optics and Photonics
• /
• v.8 no.3
• /
• pp.246-258
• /
• 2024

In aerospace and energy engineering, the reconstruction of three-dimensional (3D) temperature distributions is crucial. Traditional methods like algebraic iterative reconstruction and filtered back-projection depend on voxel division for resolution. Our algorithm, blending deep learning with computer graphics rendering, converts 2D projections into light rays for uniform sampling, using a fully connected neural network to depict the 3D temperature field. Although effective in capturing internal details, it demands multiple cameras for varied angle projections, increasing cost and computational needs. We assess the impact of camera number on reconstruction accuracy and efficiency, conducting butane-flame simulations with different camera setups (6 to 18 cameras). The results show improved accuracy with more cameras, with 12 cameras achieving optimal computational efficiency (1.263) and low error rates. Verification experiments with 9, 12, and 15 cameras, using thermocouples, confirm that the 12-camera setup as the best, balancing efficiency and accuracy. This offers a feasible, cost-effective solution for real-world applications like engine testing and environmental monitoring, improving accuracy and resource management in temperature measurement.

• Journal of Digital Convergence
• /
• v.17 no.6
• /
• pp.335-340
• /
• 2019

With the development of the media, pirate platforms and technologies are critical to the design of video content. Games based on online platforms and networking can help you develop games and expand your game as games are developed. Better yet. This paper is a major feature of the growth of the sandbox game that leads the game of InGame. Phenotypical 3D Dress Graphics You can study the usefulness of graphics based on doubles, the advantages of game characters, and space game graphics.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2003.09a
• /
• pp.50-50
• /
• 2003

World Wide Web (WWW)에서 Virtual Reality Modeling Language (VRML)를 이용하는 3 차원 (3D) 디스플레이는 사용자에게 직관적인 정보를 더 효과적으로 제공해 준다. 웹을 기반으로 하는 해부학적 영상과 융합되는 기능적 영상의 3D 가시화는 아직까지 체계적인 방식으로 연구가 활발히 진행되지 않았다. 이 연구의 목적은 2D 영상들과 함께 웹에서 VRML을 이용하여 구현되는 3D 해부학적 표면 영상들과 기능적 표면 영상들을 동시적으로 관찰할 수 있게 하고 VRML을 통해 만들어진 거리 측정 도구를 가지고 관심영역의 공간적인 위치 정보를 제공하는 것이다. 본 연구에서는 한 명의 간질 환자로부터 Magnetic Resonance (MR) 축면 영상과 발작기 및 발작간기 Single Photon Emission Computed Tomo graphy (SPECT) 축면 영상들을 각각 획득하였다. 발작 진원지의 확인을 향상시키기 위해서 subtraction ictal SPECT co registered to MRI (SISCOM) 을 수행하였다. SISCOM 결과로 나타난 각 2D 영상들은 모든 voxel 들의 평균 값 위로 1 표준편차와 2 표준편차에 해당하는 문턱 이상의 영상 값을 갖도록 하였다. SISCOM으로 나타나는 간질 발작 진원지들과 MRI 영상에서 회색질, 백색질 및 뇌척수액의 경계들을 각각 분할하고 marching cube 알고리즘에 의해 VRML 표면 영상들로 나타내었다. 축면 영상에서 실제 거리를 나타내는 x, y 축의 길이를 측정하고 z 축선의 길이를 계산하였다. VRML을 이용한 거리 측정 도구를 만들어 이전의 VRML 표면 영상들과 융합하였다. MRI 영상을 이용하여 3D 표면 영상들의 단면을 나타내고 3D 표면 영상들의 투명도를 설정하기 위해 Java Script 루틴을 사용자 인터페이스 도구로서 삽입하였다. 웹 페이지에서 구현되는 3D 표면 영상들의 투명도와 관찰 위치를 조절함에 따라 모델들 사이의 공간적인 정보를 직관적으로 알 수 있었다. 간질 발작 진원지에 대응하는 해부학적 구조를 3D 표면 영상들을 가로지르는 MRI 평면 영상들을 통해서 확인하였다. 결론적으로 본 연구에서 제시하는 웹에 근거한 3D 융합 영상의 가시화와 위치 측정은 진단 및 치료 방사선학과 외과학 등의 분야에서 온라인 방식의 연구와 교육에 있어 많은 도움을 줄 것이다.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.29 no.4 s.235
• /
• pp.570-577
• /
• 2005

In this study a complete 3D surface reconstruction method is proposed based on the concept that the vertices of surface model can be completely matched to the unstructured point cloud. In order to generate the initial mesh model from the point cloud, the mesh subdivision of bounding box and shrink-wrapping algorithm are introduced. The control mesh model for well representing the topology of point cloud is derived from the initial mesh model by using the mesh simplification technique based on the original QEM algorithm, and the parametric surface model for approximately representing the geometry of point cloud is derived by applying the local subdivision surface fitting scheme on the control mesh model. And, to reconstruct the complete matching surface model, the insertion of isolated points on the parametric surface model and the mesh optimization are carried out Especially, the fast 3D surface reconstruction is realized by introducing the voxel-based nearest-point search algorithm, and the simulation results reveal the availability of the proposed surface reconstruction method.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.535-540
• /
• 2009

Recently, autonomous robots which can achieve the complex tasks have been required with the advance of robotics. Advanced robot vision for recognition is necessary for the realization of such robots. In this paper, we propose a method to recognize an object in the actual environment. We assume that a 3D-object model used in our proposal method is the voxel data. Its inside is full up and its surface has color information. We also define the word "recognition" as the estimation of a target object's condition. This condition means the posture and the position of a target object in the actual environment. The proposal method consists of three steps. In Step 1, we extract features from the 3D-object model. In Step 2, we estimate the position of the target object. At last, we estimate the posture of the target object in Step 3. And we experiment in the actual environment. We also confirm the performance of our proposal method from results.

• Korean Journal of Digital Imaging in Medicine
• /
• v.17 no.1
• /
• pp.43-48
• /
• 2015

Intracrnial 3D TOF MR angiography was performed in 30 normal volunteers with both 1.5 and 3.0 T MRI system with high resolutions. Used Voxel sizes were $0.39{\times}0.39{\times}0.2$(1.5 T) and $0.19{\times}0.19{\times}0.35$(3.0 T), respectively. High image quality and depiction of small vessel branches were equality demonstrated with 1.5 T and 3.0 T HR TOF MRA(p<0.05). Intracranial high resolution TOF MRA with 1.5 T and 3.0 T provides high diagnostic information with having merits and demerits in depiction of vascular branches.