• 한국멀티미디어학회논문지
• /
• 제22권12호
• /
• pp.1457-1465
• /
• 2019

3D convolution is to stack multiple consecutive frames to form a cube, and then apply the 3D convolution kernel in the cube. In this structure, each feature map of the convolutional layer is connected to multiple adjacent sequential frames in the previous layer, thus capturing the motion information. However, due to the changes of pedestrian posture, motion and position, the convolution at the same place is inappropriate, and when the 3D convolution kernel is convoluted in the time domain, only time domain features of three consecutive frames can be extracted, which is not a good enough to get action information. This paper proposes an action recognition method based on feature fusion of 3D convolutional neural network. Based on the VGG16 network model, sending a pre-acquired optical flow image for learning, then get the time domain features, and then the feature of the time domain is extracted from the features extracted by the 3D convolutional neural network. Finally, the behavior classification is done by the SVM classifier.

• International Journal of Ocean System Engineering
• /
• 제1권3호
• /
• pp.136-143
• /
• 2011

In this paper, three-dimensional free-surface flows are simulated by using two different numerical methods, the constrained interpolation profile (CIP)-based and finite volume (FV)-based methods. In the CIP-based method, the governing equations are solved on stationary staggered Cartesian grids by a finite difference method, and an immersed boundary technique is applied to deal with wave-body interactions. In the FV-based method, the governing equations are solved by applying collocated finite volume discretization, and body-fitted meshes are used. A free-surface boundary is considered as the interface of the multi-phase flow with air and water, and a volumeof-fluid (VOF) approach is applied to trace the free surface. Among many variations of the VOF-type method, the tangent of hyperbola for interface capturing (THINC) and the compressive interface capturing scheme for arbitrary meshes (CICSAM) techniques are used in the CIP-based method and FV-based method, respectively. Numerical simulations have been carried out for dam-breaking and wave-body interaction problems. The computational results of the two methods are compared with experimental data and their differences are observed.

• 대한전자공학회논문지SP
• /
• 제44권1호
• /
• pp.1-9
• /
• 2007

본 논문에서는 임의 시점에서의 장면을 생성하기 위한 3차원 모델링 및 렌더링 시스템을 제안한다. 제안되는 시스템은 공간상에 설치된 복수의 스테레오 카메라와 PC들로 구성되며 UDP를 이용해 연결되어 각 카메라에서 획득되고 분석된 영상 데이터들을 모델링 PC로 전송해 실시간으로 3차원 모델을 생성하고, 이로부터 사용자가 원하는 위치에서의 장면을 생성해 디스플레이 하게 된다. 제안된 알고리듬은 성능 평가 결과 기존의 알고리듬보다 좋은 성능을 보였으며, 구현된 시스템은 실시간으로 사용자에게 원하는 시점에서의 영상을 자연스럽게 제공함을 실험을 통해 검증하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.415-416
• /
• 2006

In this paper, we proposed a new coding technique of digital hologram video using 3D scanning method and video compression technique. The proposed coding consists of capturing a digital hologram to separate into RGB color space components, localization by segmenting the fringe pattern, frequency transform using $M{\tiems}N$ (segment size) 2D DCT (2 Dimensional Discrete Cosine Transform) for extracting redundancy, 3D scan of segment to form a video sequence, motion compensated temporal filtering (MCTF) and modified video coding which uses H.264/AVC.

• Journal of Information Processing Systems
• /
• 제12권3호
• /
• pp.338-357
• /
• 2016

The recent advent of increasingly affordable and powerful 3D scanning devices capable of capturing high resolution range data about real-world objects and environments has fueled research into effective 3D surface reconstruction techniques for rendering the raw point cloud data produced by many of these devices into a form that would make it usable in a variety of application domains. This paper, therefore, provides an overview of the existing literature on surface reconstruction from 3D point clouds. It explains some of the basic surface reconstruction concepts, describes the various factors used to evaluate surface reconstruction methods, highlights some commonly encountered issues in dealing with the raw 3D point cloud data and delineates the tradeoffs between data resolution/accuracy and processing speed. It also categorizes the various techniques for this task and briefly analyzes their empirical evaluation results demarcating their advantages and disadvantages. The paper concludes with a cross-comparison of methods which have been evaluated on the same benchmark data sets along with a discussion of the overall trends reported in the literature. The objective is to provide an overview of the state of the art on surface reconstruction from point cloud data in order to facilitate and inspire further research in this area.

• 한국게임학회 논문지
• /
• 제3권2호
• /
• pp.24-29
• /
• 2003

The market of digital avatar with internet and digital technology is increasing rapidly. The users want to express any free-formed motion of their avatars in the cyber space. The user s motion capturing method as the avatar's motion can express any free-formed motion of the avatar in real-time but the methods are expensive and inconvenient. In this paper, we proposed a new method of expressing any free-formed motion of the avatar in real-time. The proposed method is an algorithm for real-time control of a 3D avatar in symmetry-formed free motion. Specially, the algorithm aims at the motion control of a 3D avatar for online dancing games. The proposed algorithm uses the skeleton character model and controls any one of two hands of the character model by a joystick with two sticks. In the symmetry-formed motion, the position and orientation of one hand can determine the position and orientation of the other hand. And the position and orientation of a hand as an end-effector can determine the pose of the arm by Inverse Kinematics. So the algorithm can control the symmetry-formed free motions of two arms by one joystick with two sticks. In the dance game, the algorithm controls the arm motion by the joystick and the other motion by the motion captured DB.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권3호
• /
• pp.1595-1613
• /
• 2017

The IR camera and laser-based IR projector provide an effective solution for real-time collection of moving targets in RGB-D videos. Different from the traditional RGB videos, the captured depth videos are not affected by the illumination variation. In this paper, we propose a novel feature extraction framework to describe human activities based on the above optical video capturing method, namely spatial-temporal texture features for 3D human activity recognition. Spatial-temporal texture feature with depth information is insensitive to illumination and occlusions, and efficient for fine-motion description. The framework of our proposed algorithm begins with video acquisition based on laser projection, video preprocessing with visual background extraction and obtains spatial-temporal key images. Then, the texture features encoded from key images are used to generate discriminative features for human activity information. The experimental results based on the different databases and practical scenarios demonstrate the effectiveness of our proposed algorithm for the large-scale data sets.

• Current Optics and Photonics
• /
• 제7권6호
• /
• pp.638-654
• /
• 2023

Holography is generally known as a technology that records and reconstructs 3D images by simultaneously capturing the intensity and phase information of light. Two or more interfering beams and illumination of this interference pattern onto a photosensitive recording medium allow us to control both the intensity and phase of light. Holography has found widespread applications not only in 3D imaging but also in manufacturing. In fact, it has been commonly used in semiconductor manufacturing, where interference light patterns are applied to photolithography, effectively reducing the half-pitch and period of line patterns, and enhancing the resolution of lithography. Moreover, holography can be used for the manufacturing of 3D regular structures (3D photonic crystals), not just surface patterns such as 1D or 2D gratings, and this can be broadly divided into (i) holographic recording and (ii) holographic lithography. In this review, we conceptually contrast two seemingly similar but fundamentally different manufacturing methods: holographic recording and holographic lithography. We comprehensively describe the differences in the manufacturing processes and the resulting structural features, as well as elucidate the distinctions in the diffractive optical properties that can be derived from them. Lastly, we aim to summarize the unique perspectives through which each method can appear distinct, with the intention of sharing information about this field with both experts and non-experts alike.

• International journal of advanced smart convergence
• /
• 제6권1호
• /
• pp.9-17
• /
• 2017

Recent development in virtual and augmented reality increases the demand for content in many different fields. One of the fast ways to create content for VR is 3D modeling of real objects. In this paper we propose a system to reconstruct three-dimensional models of real objects from the set of two-dimensional images under the assumption that the subject does not has distinct features. We explicitly consider an object that is made of one or more surfaces and radiant constant energy isotropically. We design a low cost portable multi camera rig system that is capable of capturing images simultaneously from all cameras. In order to evaluate the performance of the proposed system, comparison is made between 3D model and a CAD model. A simple algorithm is also proposed to acquire original texture or color of the subject. Using best pattern found after the experiments, 3D model of the Pyeongchang Olympic Mascot "Soohorang" is created to use as VR content.

• 대한의용생체공학회:의공학회지
• /
• 제15권2호
• /
• pp.201-208
• /
• 1994

Many three-dimensional object modeling and display methods for computer graphics and computer vision have been developed. Recently, with the help of medical imaging devices such as computerized tomography, magnetic resonance image, etc., some of those object modeling and display methods have been widely used for capturing the shape, structure and other properties of real objects in many medical applications. In this paper, we propose the reconstruction and display method of the three-dimensional object from a series of the cross sectonal image. It is implemented by using the automatic threshold selection method and the contour following algorithm. The combination of curvature and distance, we select feature points. Those feature points are the candidates for the tiling method. As a results, it is proven that this proposed method is very effective and useful in the comprehension of the object's structure. Without the technician's responce, it can be automated.