• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.35-36
• /
• 2014

본 논문에서는 Kinect 센서의 동작 인식 기능을 활용한 운동요법 앱을 설계하고 구현한다. 이 앱은 사용자의 상체, 하체 관절의 움직임을 인식하여 체조의 올바른 자세를 배울 수 있도록 한다. 이 앱의 특징은 Kinect Sensor로 인식한 관절 요소를 읽어 들이고, 각 관절의 각도를 계산하여 원하는 동작을 표현할 수 있도록 한다. 또한, 사용자의 동작과 기본 동작을 비교하여 오차 범위 내이면 새로운 동작이 진행되도록 한다. 출력되는 각 동작들은 Library를 이용하여 많은 포즈들을 입력하고 출력할 수 있기 때문에 새로운 포즈를 추가하기에 쉽다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.818-819
• /
• 2014

Recently there are many works conducted on utilizing the DIBR (Depth-Image-Based Rendering) based intermediate images for the three-dimensional displays that do not require the use of stereoscopic glasses. However the prior works have used expensive depth cameras to obtain high-resolution depth images since DIBR-based intermediate image generation method requires the accuracy for depth information. In this study, we have developed the simulation to generate multi-view intermediate images based on the depth and color images using Microsoft Kinect. This simulation aims to support the acquisition of multi-view intermediate images utilizing the low-resolution depth and color image from Kinect, and provides the integrated service for the quality evaluation of the intermediate images. This paper describes the architecture and the system implementation of this simulation program.

• Journal of the Korea Computer Graphics Society
• /
• v.22 no.4
• /
• pp.1-9
• /
• 2016

In this paper we present a projection mapping system onto human's moving hand by a projector as information delivery media and Kinect to recognize hand motion. Most traditional projection mapping techniques project a variety of images onto stationary objects, however, our system provides new user experience by projecting images onto the center of the moving palm. We explain development process of the system, and production of content as applications on our system. We propose hardware organization and development process of open software architecture based on object oriented programming approach. For stable image projection, we describe a device calibration method between the projector and Kinect in three dimensional space, and a denoising technique to minimize artifacts from Kinect coordinates vibration and unstable hand tremor.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.2
• /
• pp.45-53
• /
• 2019

RGB-D SLAM (Simultaneous Localization and Mapping) refers to the technology of using deep camera as a visual sensor for SLAM. In view of the disadvantages of high cost and indefinite scale in the construction of maps for laser sensors and traditional single and binocular cameras, a method for creating three-dimensional map of indoor environment with deep environment data combined with RGB-D SLAM scheme is studied. The method uses a mobile robot system equipped with a consumer-grade RGB-D sensor (Kinect) to acquire depth data, and then creates indoor three-dimensional point cloud maps in real time through key technologies such as positioning point generation, closed-loop detection, and map construction. The actual field experiment results show that the average error of the point cloud map created by the algorithm is 0.0045m, which ensures the stability of the construction using deep data and can accurately create real-time three-dimensional maps of indoor unknown environment.

• Journal of Broadcast Engineering
• /
• v.19 no.2
• /
• pp.140-147
• /
• 2014

In this paper, we propose a method for real-time depth image refinement. In order to improve the quality of the depth map acquired from Kinect camera, we employ constant memory and texture memory which are suitable for a 2D image processing in the graphics processing unit (GPU). In addition, we applied the joint bilateral filter (JBF) in parallel to accelerate the overall execution. To enhance the quality of the depth image, we applied the JBF hierarchically using the compute unified device architecture (CUDA). Finally, we obtain the refined depth image. Experimental results showed that the proposed real-time depth image refinement algorithm improved the subjective quality of the depth image and the computational time was 260 frames per second.

• Journal of the Society of Naval Architects of Korea
• /
• v.58 no.6
• /
• pp.382-390
• /
• 2021

The present study suggests the application of a depth camera for wave height field measurement, focusing on the calibration procedure and test setup. Azure Kinect system is used to measure the water surface elevation, with a field of view of 800 mm × 800 mm and repetition rate of 30 Hz. In the optimal optical setup, the spatial resolution of the field of view is 288 × 320 pixels. To detect the water surface by the depth camera, tracer particles that float on the water and reflects infrared is added. The calibration consists of wave height scaling and correction of the barrel distortion. A polynomial regression model of image correction is established using machine learning. The measurement results by the depth camera are compared with capacitance type wave height gauge measurement, to show good agreement.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.28 no.6
• /
• pp.599-606
• /
• 2015

International Hydrographic Organization(IHO) adopted standard S-100 as the international standard Geographic Information System(GIS) that can be generally used in the maritime sector. Accordingly, the next-generation system to support navigation information based on GIS standard technology has being developed. AR based navigation information system that supported navigation by overlapping navigation information on the CCTV image has currently being developed. In this study, we considered the application of a transparent display as a method to support efficiently this system. When a transparent display applied, the image distortion caused by using a wide-angle lens for parallax secure, and the disc s, and demonstrated the applicability of the technology by developing a prototype.

• ETRI Journal
• /
• v.45 no.6
• /
• pp.1022-1034
• /
• 2023

In this study, the impact of varying lighting conditions on recognition and decision-making was considered. The luminosity approach was presented to increase gesture recognition performance under varied lighting. An efficient framework was proposed for sensor-based sign language gesture identification, including picture acquisition, preparing data, obtaining features, and recognition. The depth images were collected using multiple Microsoft Kinect devices, and data were acquired by varying resolutions to demonstrate the idea. A case study was designed to attain acceptable accuracy in gesture recognition under variant lighting. Using American Sign Language (ASL), the dataset was created and analyzed under various lighting conditions. In ASL-based images, significant feature points were selected using the scale-invariant feature transformation (SIFT). Finally, an artificial neural network (ANN) classified hand gestures using specified characteristics for validation. The suggested method was successful across a variety of illumination conditions and different image sizes. The total effectiveness of NN architecture was shown by the 97.6% recognition accuracy rate of 26 alphabets dataset with just a 2.4% error rate.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.1
• /
• pp.113-119
• /
• 2023

In this paper, we propose an optimization of a pose estimation deep learning model for recognition of abnormal behavior in unmanned stores using radio frequencies. The radio frequency use millimeter wave in the 30 GHz to 300 GHz band. Due to the short wavelength and strong straightness, it is a frequency with less grayness and less interference due to radio absorption on the object. A millimeter wave radar is used to solve the problem of personal information infringement that may occur in conventional CCTV image-based pose estimation. Deep learning-based pose estimation models generally use convolution neural networks. The convolution neural network is a combination of convolution layers and pooling layers of different types, and there are many cases of convolution filter size, number, and convolution operations, and more cases of combining components. Therefore, it is difficult to find the structure and components of the optimal posture estimation model for input data. Compared with conventional millimeter wave-based posture estimation studies, it is possible to explore the structure and components of the optimal posture estimation model for input data using genetic algorithms, and the performance of optimizing the proposed posture estimation model is excellent. Data are collected for actual unmanned stores, and point cloud data and three-dimensional keypoint information of Kinect Azure are collected using millimeter wave radar for collapse and property damage occurring in unmanned stores. As a result of the experiment, it was confirmed that the error was moored compared to the conventional posture estimation model.