• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2010년도 추계학술대회
• /
• pp.147-148
• /
• 2010

Recently, advancement in the visual technology has lead to the further development of the three dimensional (3D) imaging systems. The visual perception to view a pair of images simultaneously, is a crucial factor to build a stereoscopic 3D image. In this paper, we present the depth cues between the intensities of the two images when viewing with both eyes. Due to this stereoscopic effect, objects at different distances from the eyes differ in their horizontal positions, giving the depth cue of horizontal disparity. By simple image processing technique, we also present the binocular disparity map between the two images. A median filter has been used to filter out all the noises occurring in the disparity map image.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권2호
• /
• pp.763-774
• /
• 2015

Hand posture recognition has been a wide region of applications in Human Computer Interaction and Computer Vision for many years. The problem arises mainly due to the high dexterity of hand and self-occlusions created in the limited view of the camera or illumination variations. To remedy these problems, a hand posture recognition method using 3-D point cloud is proposed to explicitly utilize 3-D information from depth maps in this paper. Firstly, hand region is segmented by a set of depth threshold. Next, hand image normalization will be performed to ensure that the extracted feature descriptors are scale and rotation invariant. By robustly coding and pooling 3-D facets, the proposed descriptor can effectively represent the various hand postures. After that, SVM with Gaussian kernel function is used to address the issue of posture recognition. Experimental results based on posture dataset captured by Kinect sensor (from 1 to 10) demonstrate the effectiveness of the proposed approach and the average recognition rate of our method is over 96%.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2009년도 학술대회
• /
• pp.62-67
• /
• 2009

기존의 증강현실 기반 상호작용 기법은 상호작용 주체의 원활한 추적을 위해 마커 기반의 패들이나 데이터 글러브 등의 부가적인 입력장치에 의존하거나 제한적인 3차원 입력만을 허용하는 경우가 많았다. 본 논문에서는 대상 개체 주변 공간의 물리적 점유를 감지하여 사용자의 입력을 해석하는 비접촉식 수동형 공간 상호작용 기법을 제안한다. 제안된 방법은 깊이 카메라로 구성한 증강현실 환경에 가상의 공간 센서를 배치하여 사용자의 입력을 공간적으로 해석한다. 그 결과, 부가적인 입력장치의 착용이 필요하지 않으며, 입력 형태에 대한 제약을 최소화하고, 증강된 영상 내의 상호작용 객체 간에 정확한 거리감을 제시한다. 향후 이러한 기법은 미니어쳐 AR 시스템과 같은 동적 콘텐츠 전시 플랫폼의 인터페이스에 적용될 것으로 기대된다.

• 한국게임학회 논문지
• /
• 제11권6호
• /
• pp.103-113
• /
• 2011

본 논문은 치매 예방을 위한 제스처 인식 기반 3D기능성 게임을 제안한다. 제안된 기능성 게임은 전신 제스처 인식을 이용함으로써 사용자의 두뇌 사용능력과 신체활동성을 증가시켜 치매 예방의 효과를 향상시킬 수 있도록 하였다. 기존에 개발된 제스처 인식 기술에 사용된 카메라들은 인식률과 가동영역이 한계적이다. 보다 안정적인 전신 제스처인식을 위해 3D depth 카메라로부터 사용자를 인식하고 사용자의 관절 정보를 획득하였으며 관절의 움직임을 분석하여 전신 제스처를 인식하였다. 게임 콘텐츠로는 치매의 대표적인 원인인 뇌세포의 퇴화에 초점을 맞춰 기억력, 논리력, 산술능력, 공간인지능력 등을 훈련할 수 있도록 구성하였다. 사용자 별로 게임 결과를 저장하고 분석하여 인지능력 향상 정도를 측정할 수 있도록 하였다.

• 방송공학회논문지
• /
• 제13권6호
• /
• pp.796-803
• /
• 2008

본 논문에서는 사용자에게 보다 실감있고, 사용자가 선호하는 다시점 3D 영상의 3D 입체감을 제공할 수 있는, 다시점 3D 동영상의 인식 깊이감을 조절하는 기법을 제안한다. 카메라 배열을 변경하는 기존의 방법들과 달리, 제안한 기법은 사용자가 요청하는 깊이 범위의 스케일링을 활용하기 위하여 영상의 깊이 데이터를 활용한다. 다시점 영상 및 해당 시점의 깊이 영상을 입력받아 각 화소의 변이로 변환하고, 각 화소의 변이를 조절함으로써 인식되는 깊이감을 조절하게 된다. 제안 방법은 다시점 카메라로부터 획득한 다시점 영상의 처리가 가능하며, 또한 2-시점 입체영상부터 다시점 영상까지 적용이 가능하다. 실험에서는 제안한 방법을 통하여 깊이감이 조절된 다시점 동영상을 다시점 3D 모니터로 시청하였을 때, 깊이감이 사용자가 선호하는 깊이에 따라서 스케일되는 것을 DSCQS(Double Stimulus Continuous Quality Scale)으로 확인하였다.

• Journal of the Optical Society of Korea
• /
• 제16권4호
• /
• pp.381-385
• /
• 2012

In this paper, we present a novel integral imaging pickup method. We extract each pixel's actual depth data from a real object's surface using a depth camera, then generate elemental images based on the depth map. Since the proposed method generates elemental images without a lens array, it has simplified the pickup process and overcome some disadvantages caused by a conventional optical pickup process using a lens array. As a result, we can display a three-dimensional (3D) image in integral imaging. To show the usefulness of the proposed method, an experiment is presented. Though the pickup process has been simplified in the proposed method, the experimental results reveal that it can also display a full motion parallax image the same as the image reconstructed by the conventional method. In addition, if we improve calculation speed, it will be useful in a real-time integral imaging display system.

• Journal of information and communication convergence engineering
• /
• 제13권4호
• /
• pp.275-279
• /
• 2015

There are several methods to record three dimensional (3D) information of objects such as lens array based integral imaging, synthetic aperture integral imaging (SAII), computer synthesized integral imaging (CSII), axially distributed image sensing (ADS), and axially distributed stereo image sensing (ADSS). ADSS method is capable of recording partially occluded 3D objects and reconstructing high-resolution slice plane images. In this paper, we present a computational method for depth extraction of partially occluded 3D objects using ADSS. In the proposed method, the high resolution elemental stereo image pairs are recorded by simply moving the stereo camera along the optical axis and the recorded elemental image pairs are used to reconstruct 3D slice images using the computational reconstruction algorithm. To extract depth information of partially occluded 3D object, we utilize the edge enhancement and simple block matching algorithm between two reconstructed slice image pair. To demonstrate the proposed method, we carry out the preliminary experiments and the results are presented.

• 방송공학회논문지
• /
• 제2권2호
• /
• pp.125-135
• /
• 1997

This paper is concerned with realistic mainpulation of content in video sequences. Manipulation of content in video sequences is one of the content-based functionalities for MPEG-4 Visual standard. We present an approach to synthesizing video sequences by using the intermediate outputs of three-dimensional (3D) motion and depth analysis. For concreteness, we focus on video showing 3D motion of an observer relative to a scene containing planar runways (or roads). We first present a simple runway (or road) model. Then, we describe a method of identifying the runway (or road) boundary in the image using the Point of Heading Direction (PHD) which is defined as the image of, the ray along which a camera moves. The 3D motion of the camera is obtained from one of the existing 3D analysis methods. Then, a video sequence containing a runway is manipulated by (i) coloring the scene part above a vanishing line, say blue, to show sky, (ii) filling in the occluded scene parts, and (iii) overlaying the identified runway edges and placing yellow disks in them, simulating lights. Experimental results for a real video sequence are presented.

• 한국컴퓨터정보학회논문지
• /
• 제23권9호
• /
• pp.15-20
• /
• 2018

Currently the mobile augmented reality does not require supplemental hardware, such as markers like specific images or depth cameras, in order to acquire 3D environment information. This allows for increased accessibility and usability for the general user, and also broadened scope of development. The market research enterprise Counterpoint Research predicted that the growth of the augmented reality market will reach 70 trillion Korean Won by 2021. Application developments are also progressing. Wargaming, the developer of the global game World of Tanks, has launched an application that uses AR core at the GDC 2018. Thus, we must prepare for the impending augmented reality market and acquire top class augmented reality technology using 3D environment. This paper outlines the development of a game using AR CORE SDK to acquire 3D environment and Unity 3D game engine to combine camera images and 3D graphics, which all comes together to construct a natural augmented reality.

• 로봇학회논문지
• /
• 제1권2호
• /
• pp.180-187
• /
• 2006

This paper presents a new sensor system, CALOS, for motion estimation and 3D reconstruction. The 2D laser sensor provides accurate depth information of a plane, not the whole 3D structure. On the contrary, the CCD cameras provide the projected image of whole 3D scene, not the depth of the scene. To overcome the limitations, we combine these two types of sensors, the laser sensor and the CCD cameras. We develop a motion estimation scheme appropriate for this sensor system. In the proposed scheme, the motion between two frames is estimated by using three points among the scan data and their corresponding image points, and refined by non-linear optimization. We validate the accuracy of the proposed method by 3D reconstruction using real images. The results show that the proposed system can be a practical solution for motion estimation as well as for 3D reconstruction.