• 한국멀티미디어학회논문지
• /
• 제20권11호
• /
• pp.1711-1719
• /
• 2017

In this paper, we propose a method of the zoom motion estimation for the depth video coding. The proposed method calculates the zoom ratio using the average of the depth values in the current block and in the reference block. It resizes the reference block by the zoom ratio and interpolates the reference block to size of the current block. It compares the current block with the reference block that is obtained by subtracting the average of pixels from the current block to the reference block in order to find the reference block that is the best closest one to the current block. The results of the simulation for the proposed method show that the motion estimation errors are significantly reduced.

• 한국멀티미디어학회논문지
• /
• 제16권2호
• /
• pp.131-137
• /
• 2013

동영상의 줌 움직임 추정은 구현이 아주 복잡하다. 본 논문에서는 줌 움직임 추정을 구현하기 위하여 깊이 카메라와 색상 카메라를 동시에 이용하는 방법을 제안한다. 깊이 카메라로부터 현재블록과 참조블록 사이의 거리 정보를 얻고, 이 거리 정보로부터 두 블록사이의 줌 비율을 계산한다. 줌 비율에 맞게 참조블록을 확대 또는 축소시켜 줌으로서 움직임 추정 차신호를 줄일 수 있다. 따라서, 제안된 방법은 줌 움직임 추정을 위한 복잡도가 크지 않으면서 움직임 추정 정확도를 높이는 것이 가능하다. 모의실험을 바탕으로 제안된 방법의 움직임 추정 정확도를 측정하였으며, 기존 블록정합 방법에 비하여 움직임 추정 오차값이 크게 감소함을 확인하였다.

• Earthquakes and Structures
• /
• 제6권2호
• /
• pp.141-161
• /
• 2014

This article presents a new generation of empirical ground motion models for the prediction of response spectral accelerations in soil conditions, specifically developed for the Vrancea intermediate-depth seismic source. The strong ground motion database from which the ground motion prediction model is derived consists of over 800 horizontal components of acceleration recorded from nine Vrancea intermediate-depth seismic events as well as from other seventeen intermediate-depth earthquakes produced in other seismically active regions in the world. Among the main features of the new ground motion model are the prediction of spectral ordinates values (besides the prediction of the peak ground acceleration), the extension of the magnitudes range applicability, the use of consistent metrics (epicentral distance) for this type of seismic source, the extension of the distance range applicability to 300 km, the partition of total standard deviation in intra- and inter-event standard deviations and the use of a national strong ground motion database more than two times larger than in the previous studies. The results suggest that this model is an improvement of the previous generation of ground motion prediction models and can be properly employed in the analysis of the seismic hazard of Romania.

• 대한조선학회지
• /
• 제13권1호
• /
• pp.11-16
• /
• 1976

The effect of the bow shape on the ship motion response in longitudinal regular waves of water of finite depth is investigated by employing the strip theory. The two-dimensional hydrodynamic forces(added mass and damping) were calculated by close-fit method for water of finite depth. The models for investigation are U and V bow ship forms of block coefficient 0.8 with constant after body which were used by Yourkov [2] and recently by Kim [3] for their deep water investigations. The following results are obtained by the present numerical experiments. (1) It is confirmed that the damping coefficient of the V-bow ship is greater than that of U-bow ship and in consquence the amplitude of heave and pitch of V-bow ship is smaller than that of U-bow ship among longitudinal regular head waves in water of finite depth (2) The merit of the V-bow ship on the motion damping is more significant in heave than in pitch, and is decreasing with the shallowness of water depth. (3) The change of bow form gives little effect on the wave exciting force and moment compared with the motion responce.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권5호
• /
• pp.2327-2347
• /
• 2018

Human actions can be recognized from depth sequences. In the proposed algorithm, we initially construct depth, motion maps (DMM) by projecting each depth frame onto three orthogonal Cartesian planes and add the motion energy for each view. The body part of the action (BPoA) is calculated by using bounding box with an optimal window size based on maximum spatial and temporal changes for each DMM. Furthermore, feature vector is constructed by using BPoA for each human action view. In this paper, we employed an ensemble based learning approach called Rotation Forest to recognize different actions Experimental results show that proposed method has significantly outperforms the state-of-the-art methods on Microsoft Research (MSR) Action 3D and MSR DailyActivity3D dataset.

• 한국생산제조학회지
• /
• 제24권6호
• /
• pp.605-612
• /
• 2015

Three dimensional (3D) position determination and motion recognition using a 3D depth sensor camera are applied to a developed penguin-shaped robot, and its validity and closeness are investigated. The robot is equipped with an Asus Xtion Pro Live as a 3D depth camera, and a sound module. Using the skeleton information from the motion recognition data extracted from the camera, the robot is controlled so as to follow the typical three mode-reactions formed by the operator's gestures. In this study, the extraction of skeleton joint information using the 3D depth camera is introduced, and the tracking performance of the operator's motions is explained.

• 디지털융복합연구
• /
• 제14권1호
• /
• pp.189-194
• /
• 2016

최근 카메라, 캠코더 및 CCTV 등의 사용이 활발해지면서 영상 처리 기술의 수요가 급증하고 있다. 특히 키넥트 센서와 같은 깊이(Depth) 카메라를 사용한 3D 영상 기술에 대한 연구개발이 더욱더 활성화되고 있다. 키넥트 센서는 RGB, 골격(Skeleton) 및 깊이(Depth) 영상을 통해 인체의 3D 골격 구조를 실시간 프레임 단위로 획득할 수 있는 고성능 카메라이다. 본 논문에서는 키넥트 센서를 사용하여 인체의 3D 골격 구조를 모션 캡처하고 범용으로 사용되고 있는 모션 파일 포맷($^*.trc$ 및 $^*.bvh$)으로 선택하여 저장할 수 있는 시스템을 개발한다. 또한 본 시스템은 광학식 모션 캡처 파일 포맷($^*.trc$)을 자기식 모션 캡처 파일 포맷($^*.bvh$)으로 변환할 수 있도록 하는 기능을 가진다. 마지막으로 본 논문에서는 키넥트 센서를 사용하여 캡처한 모션 데이터가 제대로 캡처되어졌는지 모션 캡처 데이터 뷰어를 통하여 확인한다.

• 한국CDE학회논문집
• /
• 제19권1호
• /
• pp.68-79
• /
• 2014

Human model simulation has been widely used in various industrial areas such as ergonomic design, product evaluation and characteristic analysis of work-related musculoskeletal disorders. However, the process of building digital human models and capturing their behaviors requires many costly and time-consuming fabrication iterations. To overcome the limitations of this expensive and time-consuming process, many studies have recently presented a markerless motion capture approach that reconstructs the time-varying skeletal motions from optical devices. However, the drawback of the markerless motion capture approach is that the phenomenon of occlusion of motion data occurs in real-time human simulation. In this study, we propose a systematic method of discriminating missing or inaccurate motion data due to motion occlusion and interpolating a sequence of motion frames captured by a markerless depth camera.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2009년도 9th International Meeting on Information Display
• /
• pp.1152-1153
• /
• 2009

Although motion impression is weak with isoluminant color stimuli, it has been shown that color signals influence motion perception. We discuss similarities and differences between color motion and luminance motion, focusing on temporal characteristics of the perception of the 2D and 3D motion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권2호
• /
• pp.1118-1133
• /
• 2017

Computer vision-based human activity recognition (HAR) has become very famous these days due to its applications in various fields such as smart home healthcare for elderly people. A video-based activity recognition system basically has many goals such as to react based on people's behavior that allows the systems to proactively assist them with their tasks. A novel approach is proposed in this work for depth video based human activity recognition using joint-based motion features of depth body shapes and Deep Belief Network (DBN). From depth video, different body parts of human activities are segmented first by means of a trained random forest. The motion features representing the magnitude and direction of each joint in next frame are extracted. Finally, the features are applied for training a DBN to be used for recognition later. The proposed HAR approach showed superior performance over conventional approaches on private and public datasets, indicating a prominent approach for practical applications in smartly controlled environments.