• Journal of Korea Game Society
• /
• v.13 no.6
• /
• pp.75-84
• /
• 2013

The recording and matching method of pose and gesture based on PC-window platform is introduced in this paper. The method uses the gesture detection camera, Xtion which is for the Windows PC. To develop the method, the API is first developed which processes and compares the depth data, RGB image data, and skeleton data obtained using the camera. The pose matching method which selectively compares only valid joints is developed. For the gesture matching, the recognition method which can differentiate the wrong pose between poses is developed. The tool which records and tests the sample data to extract the specified pose and gesture is developed. 6 different pose and gesture were captured and tested. Pose was recognized 100% and gesture was recognized 99%, so the proposed method was validated.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.4
• /
• pp.23-28
• /
• 2002

In this paper, we have proposed the hardware architecture which implements the algorithm for retaining the connectivity which prevents the disconnection in the gray-scale image thinning. To extract the skeleton from the image in a real time, it is necessary to examine the connectivity of the skeleton in a real time. The proposed architecture finds the connectivity number in the 4-clock period. The architecture consists of three blocks, PS(Parallel to Serial) Converter and Stare Generator and Ridge Checker. The PS Converter changes the 3$\times$3 gray level image to four sets of image pixels. The State Generator examines the connectivity of the central pixel by searching the data from the PS Converter. The Ridge Checker determines whether the central pixel is on the skeleton or not. The proposed architecture finds the connectivity of the central pixel in a 3$\times$3 gray level image in the 4-clocks. The total circuits are verified by the design tools and operate correctly.

• Korean Journal of Computational Design and Engineering
• /
• v.13 no.4
• /
• pp.243-254
• /
• 2008

As modular production becoming increasingly widespread in globalized manufacturing industries, sub modules or parts of the final product are being provided by many suppliers. Some part suppliers design their own products for themselves. In some cases, part suppliers provide the same type of product to multiple OEM companies. Because all part suppliers and OEM companies typically cannot use the same CAD system, engineering change in the CAD model of one company cannot be directly propagated to related CAD models of other companies. Even if two companies use the same CAD system, it may be difficult to share their CAD model owing to corporate security policy. In this paper, a neutral reference model that consists of a neutral skeleton model and an external reference data model is proposed as a new medium for the sharing and propagation of engineering change information among collaborating companies.

• IEIE Transactions on Smart Processing and Computing
• /
• v.1 no.1
• /
• pp.1-7
• /
• 2012

This paper presents a system used to automatically recognize the road traffic control gestures of police officers. In this approach,the control gestures of traffic police officers are captured in the form of depth images.A human skeleton is then constructed using a kinematic model. The feature vector describing a traffic control gesture is built from the relative angles found amongst the joints of the constructed human skeleton. We utilize Support Vector Machines (SVMs) to perform the gesture recognition. Experiments show that our proposed method is robust and efficient and is suitable for real-time application. We also present a testbed system based on the SVMs trained data for real-time traffic gesture recognition.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.12
• /
• pp.2314-2318
• /
• 2010

For human-robot interaction, a robot should recognize the meaning of human behavior. In the case of static behavior such as face expression and sign language, the information contained in a single image is sufficient to deliver the meaning to the robot. In the case of dynamic behavior such as gestures, however, the information of sequential images is required. This paper proposes behavior classification by using fuzzy classifier to deliver the meaning of dynamic behavior to the robot. The proposed method extracts feature points from input images by a skeleton model, generates a vector space from a differential image of the extracted feature points, and uses this information as the learning data for fuzzy classifier. Finally, we show the effectiveness and the feasibility of the proposed method through experiments.

• The korean journal of orthodontics
• /
• v.9 no.1
• /
• pp.67-84
• /
• 1979

The author measured the right and left craniofacial skeleton using posteroanterior cephalography of 124 male and 97 female subjects of ages ranging 2-6 years, 7-13 years and adult group on Korean. Statstical parameters were calculated by computer from the data. The results of the study were as follows. 1. In the male and female group of deciduous dentition, the right side measurements were slightly greater than the left side measurements, and generally male measurements were greater than female measurements. 2. In the both sex group of mixed dentition, the right side measurements were greater than the left side measurements, and in comparision of sex, the female recognized more growth, and development than male, especially on the cranial base, lateral maxilla, lower maxilla and dental region. 3. In the permanent dentition group, regardless of sex, the right side measurements were greater than the left side measurements, and in the sex difference, male was conspicuously greater than female on both sides measurements. 4. In the permanent dentition group, the asymmetry of the face was striking than deciduous dentition and mixed dentition group.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2004.11a
• /
• pp.301-306
• /
• 2004

The purpose of this study is to supply basic data for distinction of Support and Infill in Open Housing. This study is based on archive survey and actual investigation through visiting four sites in Japan, which are experimental SI(skeleton & infill) housings. We found a lot of distinction techniques and methods applied to the SI housings, and findings are applicable to the planning and designing works of apartment housings. Based on the theoretical Studies concerning the distinction and separation of SI housing, this study made out interface map for separation and distinction of housing components and examined a design method of separation accordingly.

• Geomechanics and Engineering
• /
• v.3 no.2
• /
• pp.83-108
• /
• 2011

The effect of non-plastic fines (silt) on liquefaction and pore pressure generation characteristics of saturated sands was studied through undrained stress controlled cyclic triaxial tests using cylindrical specimens of size 50 mm diameter and height 100 mm at different cyclic stress ratios and at a frequency of 0.1 Hz. The tests were carried out in the laboratory adopting various measures of sample density through various approaches namely gross void ratio approach, relative density approach, sand skeleton void ratio approach, and interfine void ratio approach. The limiting silt content and the relative density of a specimen were found to influence the undrained cyclic response of sand-silt mixtures to a great extent. Undrained cyclic response was observed to be independent of silt content at very high relative densities. However, the presence of fines significantly influenced this response of loose to medium dense specimens. Combined analyses of cyclic resistance have been done using the entire data collected from all the approaches.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.78-79
• /
• 2022

In this paper, we propose a technique for tracking the classification and location of each user through body proportion analysis of the normalized skeletons of multiple users obtained using RGB-D cameras. To this end, each user's 3D skeleton is extracted from the 3D point cloud and body proportion information is stored. After that, the stored body proportion information is compared with the body proportion data output from the entire frame to propose a user classification and location tracking algorithm in the entire image.

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.592-593
• /
• 2022

다중 데이터 기반의 행동 인식 과정에서 데이터 수집 반경이 비교적 제한되는 영상 데이터의 결측에 대한 보완이 요구된다. 본 논문에서는 6축 센서 데이터를 이용하여 결측된 영상 데이터를 생성함으로써 행동 인식의 성능을 개선하는 방법을 제안한다. 가속도와 자이로 센서로부터 수집된 행동 데이터를 이용하여 GAN(Generative Adversarial Network)을 통해 영상에서의 관절(Skeleton) 움직임에 대한 데이터를 생성하고자 한다. 이를 위해 DeepLabCut 기반 모델 학습을 통해 관절 좌표를 추출하며, 전처리된 센서 시퀀스 데이터를 가지고 GRU 기반 GAN 모델을 통해 관절 좌표에 대한 영상 시퀀스 데이터를 생성한다. 생성된 영상 시퀀스 데이터는 영상 데이터의 결측이 발생했을 때 대신 행동 인식 모델의 입력값으로 활용될 수 있어 성능 향상을 기대할 수 있다.