• Journal of Korea Multimedia Society
• /
• v.12 no.9
• /
• pp.1217-1232
• /
• 2009

In this paper, we propose a semi-automatic 3D building reconstruction method using uncalibrated images which includes the facade of target building. First, we extract feature points in all images and find corresponding points between each pair of images. Second, we extract lines on each image and estimate the vanishing points. Extracted lines are grouped with respect to their corresponding vanishing points. The adjacency graph is used to organize the image sequence based on the number of corresponding points between image pairs and camera calibration is performed. The initial solid model can be generated by some user interactions using grouped lines and camera pose information. From initial solid model, a detailed building model is reconstructed by a combination of predefined basic Euler operators on half-edge data structure. Automatically computed geometric information is visualized to help user's interaction during the detail modeling process. The proposed system allow the user to get a 3D building model with less user interaction by augmenting various automatically generated geometric information.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.508-510
• /
• 2001

3차원 물체의 인식과 공간 정보를 추출해 내는 것이 물체인식의 주요 목적이다. 본 논문에서는 평면의 표면을 갖는 기하학적 물체들을 인식하는데 인공신경망이 적용 가능함이 조사되었다. 물체인식을 위한 모델들은 CAD모델들로부터 자동적으로 추출되며, 획득된 물체의 영상과 일치하는 물체의 국면(aspect)과의 매칭은 조건만족 인경신경망을 이용하여 매칭-오차를 최소화시키는 방법을 처리되었다. 인식된 물체의 국면이 어느 방향에서 획득되었는지에 대한 정보(Aspect's view direction)는 검색된 가시 평면들의 분포로부터 추출됨을 ART와 같은 인공신경망을 이용하여 실시간으로 복원할 수 있음을 보였다. 대표적이 측정방향과 이 측정방향으로부터의 편차들을 한 범주에 넣고 학습을 통해 정확한 측정방향 정보들을 구하며, 획득된 3차원 물체의 영상들에 따라 자동적으로 측정방향범주 들이 추가되도록 한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.185-188
• /
• 2015

미리 저장된 객체의 3차원 특징점(Feature point) 좌표와 카메라 영상의 2차원 특징점 좌표를 매칭(Matching)하여 객체를 추적하는 방식의 경우, 카메라의 시점이 변할 때 특징점에서 발생되는 원근 효과(Perspective effect)가 반영되지 못하여 특징점 매칭 오류가 발생한다. 따라서 특징점에서 발생하는 원근 효과를 반영하여 정확한 카메라 포즈를 추정하기 위해 이전 프레임(Frame)의 카메라 포즈(Camera Pose)에 맞추어 텍스쳐가 포함 된 3차원 객체의 모델을 렌더링 하여 원근 효과를 적용한 후, 현재 카메라 영상과 특징점 매칭하여 프레임 사이의 카메라 움직임을 구하여 객체를 추적한다. 더 나아가 본 논문에서는 특징점 매칭에서 발생하는 작은 오류들로 인한 미세한 카메라 움직임은 2단계의 임계치(Threshold)를 적용하여 떨림 현상으로 간주하여 떨림 현상이 제거된 객체 추적을 수행한다. 매 프레임마다 카메라 포즈에 맞춘 추적 객체를 렌더링 하기 때문에 떨림 현상으로 간주되어 제거된 카메라 움직임은 누적되지 않고, 추적 오류도 발생시키지 않는다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.667-669
• /
• 2022

2차원 이미지를 통한 자세 추정의 경우 관절이 겹치거나 가려져 있는 등의 인식 저해 요소로 인하여 자세 추정 정확도가 감소하는 한계가 있다. 본 논문에서는 GAN을 통해 2차원 이미지를 3차원으로 증강한 뒤 자세를 추정하는 기법을 제안한다. 제안하는 방법은 2차원 이미지의 평면좌표 값에서 GAN을 통해 노이즈 벡터 z축 값과 피사체에 투영되는 빛의 방향 값을 반영한 3차원 이미지를 만든다. 이러한 이미지 합성 과정을 거친 후 DeepLabCut을 사용해 관절 좌표를 추출하고 자세 추정 및 분류를 진행한다. 이를 통해 2차원에서의 자세 추정 정확도 향상을 기대할 수 있으며, 향후 이를 기반한 이상행동 탐지 분야에서 적용할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.600-602
• /
• 2022

2차원 이미지 또는 영상을 통한 자세 추정의 경우, 영상 내에서 발생할 수 있는 탐지 오류, 피사체 잘림, 폐색(Occlusion) 등으로 인해 자세 추정 정확도가 감소할 수 있다. 본 논문에서는 4장 이상의 다양한 각도로 촬영한 이미지를 NeRF(Neural Radiance Fields)를 통해 이미지 합성(Image synthesis)을 진행하여 3차원 모델을 생성한다. 이후 DeepLabCut을 사용하여 관절 좌표와 골격(Skeleton)을 구축한다. 구축한 골격을 인공지능에 학습시킨 뒤 2차원 영상에서의 관절 좌표 인식, 골격 구축, 자세 추정을 진행한다. 2차원 영상 테스트 데이터를 통해, 3차원 모델을 사전 학습한 인공지능 모델과 기존 2차원 이미지를 사용하여 학습한 인공지능 모델의 자세 추정 정확도를 비교한다.

• The Korean Journal of Food And Nutrition
• /
• v.15 no.1
• /
• pp.64-69
• /
• 2002

To know a degree of pollution in agricultural products cultivated around 4 unworked mines located in Chollanam-Do, we investigated a concentration of heavy metal and trace metal to provide the basic data for its residual limits. 28 samples of seven kinds of agricultural products(lettuce, onion, potato, radish, pepper, pumpkin, bean) collected from 4 unworked mines located in Chollanam-Do in 2001 were analyzed by Mercury Analyzer for mercury and Atomic Absorption spectrophotometer for Pb, As, Cd, Cu, Mn, and Zn. Hg contents were detected N.D. ∼trace level(0.01 ppm and less) and As content were detected N.D. ∼ 0.029 lpm but most of same)leE were not detected. Cd contents were detected N.D∼0.124 ppm. The results of Hg, As, Pb and Cd content showed that for all the 7 species of agricultural products studied, none have accumulated levels dangerous enough to Pose health problems. The average contents of Cu were 3.070 ∼ 7.825 ppm in bean, the Mn were 3.688 ∼23.935 ppm in lettuce ailed the Zn were 5.690 ∼21.171 ppm in bean, respectively.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.12
• /
• pp.1259-1266
• /
• 2014

Road signs provide important safety information about road and traffic conditions to drivers. Road signs include not only common traffic signs but also warning information regarding unexpected obstacles and road constructions. Therefore, accurate detection and identification of road signs is one of the most important research topics related to safe driving. In this paper, we propose a 3-D vision technique to automatically detect and track road signs in a video sequence which is acquired from a stereo vision camera mounted on a vehicle. First, color information is used to initially detect the sign candidates. Second, the SVM (Support Vector Machine) is employed to determine true signs from the candidates. Once a road sign is detected in a video frame, it is continuously tracked from the next frame until it is disappeared. The 2-D position of a detected sign in the next frame is predicted by the 3-D motion of the vehicle. Here, the 3-D vehicle motion is acquired by using the 3-D pose information of the detected sign. Finally, the predicted 2-D position is corrected by template-matching of the scaled template of the detected sign within a window area around the predicted position. Experimental results show that the proposed method can detect and track many types of road signs successfully. Tracking comparisons with two different methods are shown.

• Journal of Korea Multimedia Society
• /
• v.11 no.11
• /
• pp.1575-1591
• /
• 2008

In this paper, we propose a new method detecting foot motion capture in order to overlap in realtime foot's 3D virtual model from stereo cameras. In order to overlap foot's virtual model at the same position of the foot, a process of the foot's joint detection to regularly track the foot's joint motion is necessary, and accurate register both foot's virtual model and user's foot in complicated motion is most important problem in this technology. In this paper, we propose a dynamic registration using two types of marker groups. A plane information of the ground handles the relationship between foot's virtual model and user's foot and obtains foot's pose and location. Foot's rotation is predicted by two attached marker groups according to instep of center framework. Consequently, we had implemented our proposed system and estimated the accuracy of the proposed method using various experiments.

• Journal of Korea Multimedia Society
• /
• v.15 no.8
• /
• pp.963-976
• /
• 2012

'Gesture' except for voice is the most intuitive means of communication. Thus, many researches on how to control computer using gesture are in progress. User detection and tracking in these studies is one of the most important processes. Conventional 2D object detection and tracking methods are sensitive to changes in the environment or lights, and a mix of 2D and 3D information methods has the disadvantage of a lot of computational complexity. In addition, using conventional 3D information methods can not segment similar depth object. In this paper, we propose object detection and tracking method using Depth Projection Map that is the cumulative value of the depth and motion information. Simulation results show that our method is robust to changes in lighting or environment, and has faster operation speed, and can work well for detection and tracking of similar depth objects.

• Aerospace Engineering and Technology
• /
• v.12 no.1
• /
• pp.128-136
• /
• 2013

Recently, the localization recognition system research has been studied using various sensors according to increased interest in autonomous navigation flight. In case of indoor environment which cannot support GPS information, we have to look for another way to recognize current position. The Image-based localization recognition system has been interested although there are lots of way to know current pose. In this paper, we explain the localization recognition system based on mark and implementation of autonomous navigation flight. In order to apply to real environment which cannot support marks, localization based on real-time 3D map building is discussed.