• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.510-511
• /
• 2017

In this paper, a study was conducted to prevent spinal - related diseases and to help posture correction by feeding back the wrong attitude to the users. Kinect sensor was used for this purpose. In order to measure the movement of the user, the degree of motion change was measured by indexing the skeletal data coordinate value. It is confirmed that the implemented system can observe not only posture but also distraction of user.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.07a
• /
• pp.262-263
• /
• 2015

본 논문에서는 마이크로소프트 키넥트 센서를 이용한 실시간 성능의 3차원 환경 복원 알고리즘을 소개한다. 기존의 멀티키넥트 시스템을 확장하여 총 열두대의 키넥트를 사용하는데, 키넥트의 높은 대역폭 문제를 해결하기 위하여 키넥트가 여섯 대씩 연결된 두 대의 데스크탑을 UDP 통신으로 연결하였고, 각 키넥트로부터 들어오는 3차원 포인트클라우드로부터 확률적인 3차원 환경복원을 하기 위하여 옥토맵 알고리즘을 차용하였다. 또한, GPU를 연산에 활용함으로써 실시간 성능을 확보하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.605-606
• /
• 2011

최근 PC 기반의 새로운 입력장치인 키넥트(Kinect) 센서에 대한 연구사 활발하게 진행되고 있다. 키넥트 센서의 가장 큰 이점은 2차원 평면 영상정보에 대하여 새롭게 추가된 깊이(Depth) 정보를 얻을 수 있다는 것이다. 이것은 이전에 등장했던 다른 인터페이스보다 새로운 차원의 인터페이스로서 2차원의 영상 정보로부터 한 차원 확장된 3차원의 정보를 활용할 수 있다는 점에서 그 의미가 크다고 볼 수 있다. 본 논문에서는 이러한 2차원 영상정보와 추가된 깊이 정보를 활용하여 사람의 영상정보와 사람의 위치 정보를 활용하여 위치한 환경의 조명을 제어할 수 있도록 시스템을 구성하였다. 또한, 키넥트의 다양한 성능을 검토하기위하여 3개의 키넥트를 사용하였으며 중첩된 영상정보를 30%의 중복도를 가지도록 구성하여 다양한 활용가능성을 검토하였다. 구현한 시스템을 통하여 사람을 추적할 수 있도록 알고리즘을 개발하고, 추적된 사람의 위치 정보를 통하여 LED 조명을 제어할 수 있는 키넥트 기반의 조명제어 시스템을 제시하고자 한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06c
• /
• pp.448-450
• /
• 2012

본 논문에서는 Kinect 방식의 RGB-D 영상센서를 사용하여, 깊이(Depth) 영상으로부터 3차원 객체의 기하정보를 표현하는 표면 정규 벡터(Surface Normal Vector)를 추출하고, 그 결과를 영상화하는 방법을 제안하며, 제안된 방법으로 생성된 영상으로부터 깊이 영상의 특징점 및 특징 기술자를 추출하여 3차원 객체 인식 성능을 향상시키는 방법을 제안한다. 또한 생성된 RGB-D 특징 기술자들을 객체 단위로 구분 가능한 코드북(CodeBook) 학습을 통한 인식방법을 제안하여 객체의 인식 성능을 높이는 방법을 제안한다. 제안하는 RGB-D 기반의 특징 추출 및 학습 방법은 텍스쳐 유무, 카메라 회전 및 이동 변화 등의 환경변화에 강건함을 실험적으로 증명하였으며, 이 방법은 Kinect 방식의 RGB-D 영상을 사용하는 3차원 객체/공간 인식 및 추적, 혹은 이를 응용하는 증강현실 시스템에 적용하여 사용될 수 있다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.137-138
• /
• 2019

최근에는 청소년과 어른 대상으로한 운동을 많이 함으로써 운동은 그 어느 때 보다 인기를 얻고 있다. 하지만 그에 따른 사고도 점차 많이 일어나고 있다. 이는 신체활동 능력의 높이는 동시에 운동자세교정을 해주고 사고방지를 하고자 스포츠, 교육, 3D 가상현실 일부 사용되는 몰입형 프로그램으로 설계 했다. 본 프로그램에서는 사용자 동작을 효과적으로 인식 할 수 있는 키넥트(Kinect)를 사용하여 인터페이스를 제작하였고 향후 연구를 통해 운동자세효과를 검증하고 다양한 운동 그리고 재활치료등 개발할 필요가 있음을 시사한다.

• International Journal of Computer Science & Network Security
• /
• v.22 no.6
• /
• pp.97-108
• /
• 2022

The video-assisted human action recognition [1] field is one of the most active ones in computer vision research. Since the depth data [2] obtained by Kinect cameras has more benefits than traditional RGB data, research on human action detection has recently increased because of the Kinect camera. We conducted a systematic study of strategies for recognizing human activity based on deep data in this article. All methods are grouped into deep map tactics and skeleton tactics. A comparison of some of the more traditional strategies is also covered. We then examined the specifics of different depth behavior databases and provided a straightforward distinction between them. We address the advantages and disadvantages of depth and skeleton-based techniques in this discussion.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.10
• /
• pp.920-927
• /
• 2013

Chroma key technique which composes images by separating an object from its background in specific color has restrictions on color and space. Especially, unlike general chroma key technique, image composition for stereo 3D display requires natural image composition method in 3D space. The thesis attempted to compose images in 3D space using depth keying method which uses high resolution depth information. High resolution depth map was obtained through camera calibration between the DSLR and Kinect sensor. 3D mesh model was created by the high resolution depth information and mapped with RGB color value. Object was converted into point cloud type in 3D space after separating it from its background according to depth information. The image in which 3D virtual background and object are composed obtained and played stereo 3D images using a virtual camera.

• Journal of Korea Game Society
• /
• v.14 no.1
• /
• pp.49-58
• /
• 2014

We have developed a serious game for the elderly named 'Paldokangsan3' as a next version of 'Paldokangsan2' which was a walking game with memorization in shopping streets using controllers of hand buttons and foot boards with PC developed in 2012. The interface of had buttons and food boards were changed into the interface of motion capture with the aid of Kinect. The expected effects are improvement of memorization, improvement of concentration ability to find out items in the vision, improvement of vitality through physical walking actions, as well as improvement of mental health through entertainment activities of game play. We have tested 25 volunteers of more than 65 years old in Cheonan city community center for the elderly for 2 weeks. We got the result that the game play was showing positive effectiveness on memorization, physical exercise and self-esteem in mental health area of the elderly.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.4
• /
• pp.127-132
• /
• 2014

In this paper, we propose a digital signage method that can control the contents of the screen and download the desired files to individual smart terminal with no touch of digital signage screen by using Kinect. The main features and contributions of the proposed method are as follows. First, can solve the problems such as the risk of bacterial infection, not cleanliness on the screen as well as maximize advertising effectiveness. Second, can support bi-directional communications with computer while we touch only wanted advertisements. The performance evaluation shows the effectiveness of the proposed method. Especially, we can see that the proposed system operates through the recognition of user's actions when the cursor is within the scope of a certain coordinate and the file transfer time to download information is different depending on the PC performance and the version of personal smart device.

• KIPS Transactions on Computer and Communication Systems
• /
• v.5 no.10
• /
• pp.303-310
• /
• 2016

Recently there has been growing popularity in exergame, such as Wii Sport or Xbox Fitness game, which enables users to get physical exercise while playing the games. In such experienced exercise games, the user's posture recognition is very important to find out exactly how much the users need to take their body posture as compared to the proper posture. This paper proposes a new exercise posture recognition model designed for the exercise game content for the elderly. The proposed model is based on extracting feature points of a skeleton model provided by the Kinect sensor to generate the feature vectors to recognize the user's exercise posture information. This paper describes the design and implementation of the exercise posture recognition model and demonstrates the feasibility of this proposed posture recognition model through a simple experiment. The experimental results showed 94.52% of average accordance rate for 12 exercise postures of 10 participants.