• Journal of Korea Multimedia Society
• /
• v.22 no.12
• /
• 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.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.409-419
• /
• 2023

Structural integrity can be accessed from dynamic deformations of structures. Moreover, dynamic deformations can be acquired from non-contact sensors such as video cameras. Kanade-Lucas-Tomasi (KLT) algorithm is one of the commonly used methods for motion tracking. However, averaging throughout the extracted features would induce bias in the measurement. In addition, pixel-wise measurements can be converted to physical units through camera intrinsic. Still, the depth information is unreachable without prior knowledge of the space information. The assigned homogeneous coordinates would then mismatch manually selected feature points, resulting in measurement errors during coordinate transformation. In this study, a two-stage optimization method for video-based measurements is proposed. The manually selected feature points are first optimized by minimizing the errors compared with the homogeneous coordinate. Then, the optimized points are utilized for the KLT algorithm to extract displacements through inverse projection. Two additional criteria are employed to eliminate outliers from KLT, resulting in more reliable displacement responses. The second-stage optimization subsequently fine-tunes the geometry of the selected coordinates. The optimization process also considers the number of interpolation points at different depths of an image to reduce the effect of out-of-plane motions. As a result, the proposed method is numerically investigated by using a truss bridge as a physics-based graphic model (PBGM) to extract high-accuracy displacements from recorded videos under various capturing angles and structural conditions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1794-1806
• /
• 2023

This study presents a method for capturing photographs of users as input and converting them into 2D character animation sprites using a generative adversarial network-based artificial intelligence network. Traditionally, 2D character animations have been created by manually creating an entire sequence of sprite images, which incurs high development costs. To address this issue, this study proposes a technique that combines motion videos and sample 2D images. In the 2D sprite generation process that uses the proposed technique, a sequence of images is extracted from real-life images captured by the user, and these are combined with character images from within the game. Our research aims to leverage cutting-edge deep learning-based image manipulation techniques, such as the GAN-based motion transfer network (impersonator) and background noise removal (U² -Net), to generate a sequence of animation sprites from a single image. The proposed technique enables the creation of diverse animations and motions just one image. By utilizing these advancements, we focus on enhancing productivity in the game and animation industry through improved efficiency and streamlined production processes. By employing state-of-the-art techniques, our research enables the generation of 2D sprite images with various motions, offering significant potential for boosting productivity and creativity in the industry.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.6
• /
• pp.181-190
• /
• 2021

Sports activities conducted by multiple people are difficult to proceed in a situation where a widespread epidemic such as COVID-19 is spreading, and this causes a lack of physical activity in modern people. This problem can be overcome by using online exercise contents, but it is difficult to check detailed postures such as during face-to-face exercise. In this study, we present a model that detects posture and tracks movement using IT system for better non-face-to-face exercise content management. The proposed motion tracking model defines a body model with reference to motion analysis methods widely used in physical education and defines posture and movement accordingly. Using the proposed model, it is possible to recognize and analyze movements used in exercise, know the number of specific movements in the exercise program, and detect whether or not the exercise program is performed. In order to verify the validity of the proposed model, we implemented motion tracking and exercise program tracking programs using Azure Kinect DK, a markerless motion capture device. If the proposed motion tracking model is improved and the performance of the motion capture system is improved, more detailed motion analysis is possible and the number of types of motions can be increased.

• Steel and Composite Structures
• /
• v.20 no.5
• /
• pp.963-981
• /
• 2016

A nonlocal trigonometric shear deformation beam theory based on neutral surface position is developed for bending, buckling, and vibration of functionally graded (FG) nanobeams using the nonlocal differential constitutive relations of Eringen. The present model is capable of capturing both small scale effect and transverse shear deformation effects of FG nanobeams, and does not require shear correction factors. The material properties of the FG nanobeam are assumed to vary in the thickness direction. The equations of motion are derived by employing Hamilton's principle, and the physical neutral surface concept. Analytical solutions are presented for a simply supported FG nanobeam, and the obtained results compare well with those predicted by the nonlocal Timoshenko beam theory.

• Journal of Magnetics
• /
• v.8 no.1
• /
• pp.70-73
• /
• 2003

We have developed a new tracking system of jaw movement. The system consists of two permanent NdFeB magnets and 32 elements of two-axial fluxgate sensor array, The two magnets are attached to head portion and front tooth. This system does not need any attachments of the head portion or mouth such as clutch or magnetic field sensor except magnets. The proposed system is applicable for five degree of freedom. Position accuracy within 2]m was achieved. We developed a 3D computer input device by using the above mentioned technique.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• 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.

• Proceedings of the Korean Magnestics Society Conference
• /
• 2002.12a
• /
• pp.12-12
• /
• 2002

We present the new jaw tracking system that utilizes five degrees of freedom of head and lower jaw. The proposed method does not disturb the physiology because the system does not need to fix magnetic field sensors or clutch on the patient's body or mouth. The impact of this result goes beyond the conventional optical and magnetic tracking system of jaw movement [1][2]. Position accuracy within 1mm and reasonable cost is necessary for practical use of the jaw tracking system. (omitted)

• Advances in concrete construction
• /
• v.13 no.5
• /
• pp.361-365
• /
• 2022

The objective of this study is to simulate the two-phase flow in pipes with various two-fluid models and determinate the shear stress. A hyperbolic shear deformation theory is used for modelling of the pipe. Two-fluid models are solved by using the conservative shock capturing method. Energy relations are used for deriving the motion equations. When the initial conditions of problem satisfied the Kelvin Helmholtz instability conditions, the free-pressure two-fluid model could accurately predict discontinuities in the solution field. A numerical solution is applied for computing the shear stress. The two-pressure two-fluid model produces more numerical diffusion compared to the free-pressure two-fluid and single-pressure two-fluid models. Results show that with increasing the two-phase percent, the shear stress is reduced.

• Journal of The Korean Astronomical Society
• /
• v.34 no.4
• /
• pp.191-201
• /
• 2001

My contribution to these proceedings summarizes a general overview on High Resolution Shock Capturing methods (HRSC) in the field of relativistic hydrodynamics with special emphasis on Riemann solvers. HRSC techniques achieve highly accurate numerical approximations (formally second order or better) in smooth regions of the flow, and capture the motion of unresolved steep gradients without creating spurious oscillations. In the first part I will show how these techniques have been extended to relativistic hydrodynamics, making it possible to explore some challenging astrophysical scenarios. I will review recent literature concerning the main properties of different special relativistic Riemann solvers, and discuss several 1D and 2D test problems which are commonly used to evaluate the performance of numerical methods in relativistic hydrodynamics. In the second part I will illustrate the use of HRSC methods in several astrophysical applications where special and general relativistic hydrodynamical processes play a crucial role.