• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.667-670
• /
• 2020

최근에는 증강현실, 로봇공학 등의 분야에서 객체의 위치 검출 이외에도, 객체의 자세에 대한 추정이 요구되고 있다. 객체의 자세 정보가 포함된 데이터셋은 위치 정보만 포함된 데이터셋에 비하여 상대적으로 매우 적기 때문에 인공 신경망 구조를 활용하기 어려운 측면이 있으나, 최근에 들어서는 기계학습 기반의 자세 추정 알고리즘들이 여럿 등장하고 있다. 본 논문에서는 이 가운데 Dense 6d Pose Object detector (DPOD) [11]의 구조를 기반으로 하여 가구의 조립 설명서에 그려진 가구 부품들의 자세를 추정하고자 한다. DPOD [11]는 입력으로 RGB 영상을 받으며, 해당 영상에서 자세를 추정하고자 하는 객체의 영역에 해당하는 픽셀들을 추정하고, 객체의 영역에 해당되는 각 픽셀에서 해당 객체의 3D 모델의 UV map 값을 추정한다. 이렇게 픽셀 개수만큼의 2D - 3D 대응이 생성된 이후에는, RANSAC과 PnP 알고리즘을 통해 RGB 영상에서의 객체와 객체의 3D 모델 간의 변환 관계 행렬이 구해지게 된다. 본 논문에서는 사전에 정해진 24개의 자세 후보들을 기반으로 가구 부품의 3D 모델을 2D에 투영한 RGB 영상들로 인공 신경망을 학습하였으며, 평가 시에는 실제 조립 설명서에서의 가구 부품의 자세를 추정하였다. 실험 결과 IKEA의 Stefan 의자 조립 설명서에 대하여 100%의 ADD score를 얻었으며, 추정 자세가 자세 후보군 중 정답 자세에 가장 근접한 경우를 정답으로 평가했을 때 100%의 정답률을 얻었다. 제안하는 신경망을 사용하였을 때, 가구 조립 설명서에서 가구 부품의 위치를 찾는 객체 검출기(object detection network)와, 각 개체의 종류를 구분하는 객체 리트리벌 네트워크(retrieval network)를 함께 사용하여 최종적으로 가구 부품의 자세를 추정할 수 있다.

• The Journal of Information Technology
• /
• v.7 no.2
• /
• pp.93-104
• /
• 2004

Sports scene is characterized by large amount of global motion due to pan and zoom of camera motion, and includes many small objects moving independently. Some short period of sports games is thrilling to televiewers, and important to producers. At the same time that kinds of scenes exhibit exceptionally dynamic motions and it is very difficult to analyze the motions with conventional algorithms. In this thesis, several algorithms are proposed for global motion analysis on these dynamic scenes. It is shown that proposed algorithms worked well for motion compensation and panorama synthesis. When cascading the inter frame motions, accumulated errors are unavoidable. In order to minimize these errors, interpolation method of motion vectors is introduced. Affined transform or perspective projection transform is regarded as a square matrix, which can be factorized into small amount of motion vectors. To solve factorization problem, we preposed the adaptation of Newton Raphson method into vector and matrix form, which is also computationally efficient. Combining multi frame motion estimation and the corresponding interpolation in hierarchical manner enhancement algorithm of motion parameters is proposed, which is suitable for motion compensation and panorama synthesis. The proposed algorithms are suitable for special effect rendering for broadcast system, video indexing, tracking in complex scenes, and other fields requiring global motion estimation.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.11
• /
• pp.47-55
• /
• 2012

In this paper, we propose a method to reconstruct accurate 3D shape object by using multi-view images which are disturbed by motion blur. In multi-view deblurring, more precise PSF estimation can be done by using the geometric relationship between multi-view images. The proposed method first estimates initial 2D PSFs from individual input images. Then 3D PSF candidates are projected on the input images one by one to find the best one which are mostly consistent with the initial 2D PSFs. 3D PSF consists with direction and density and it represents the 3D trajectory of object motion. 야to restore 3D shape by using multi-view images computes the similarity map and estimates the position of 3D point. The estimated 3D PSF is again projected to input images and they replaces the intial 2D PSFs which are finally used in image deblurring. Experimental result shows that the quality of image deblurring and 3D reconstruction improves significantly compared with the result when the input images are independently deblurred.

• The Journal of the Korea Contents Association
• /
• v.7 no.2
• /
• pp.117-124
• /
• 2007

This paper describes methodology that enables animators to create the facial expression animations and to control the facial expressions in real-time by reusing motion capture datas. In order to achieve this, we fix a facial expression state expression method to express facial states based on facial motion data. In addition, by distributing facial expressions into intuitive space using LLE algorithm, it is possible to create the animations or to control the expressions in real-time from facial expression space using user interface. In this paper, approximately 2400 facial expression frames are used to generate facial expression space. In addition, by navigating facial expression space projected on the 2-dimensional plane, it is possible to create the animations or to control the expressions of 3-dimensional avatars in real-time by selecting a series of expressions from facial expression space. In order to distribute approximately 2400 facial expression data into intuitional space, there is need to represents the state of each expressions from facial expression frames. In order to achieve this, the distance matrix that presents the distances between pairs of feature points on the faces, is used. In order to distribute this datas, LLE algorithm is used for visualization in 2-dimensional plane. Animators are told to control facial expressions or to create animations when using the user interface of this system. This paper evaluates the results of the experiment.

• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.67-75
• /
• 2018

Multidimensional scaling (MDS) is an exploratory analysis of multivariate data to represent the dissimilarity among objects in the geometric low-dimensional space. However, a general MDS map only shows the information of objects without any information about variables. In this study, we used MDS based on the algorithm of Torgerson (Theory and Methods of Scaling, Wiley, 1958) to visualize some clusters of objects in categorical data. For this, we convert given data into a multiple indicator matrix. Additionally, we added the information of levels for each categorical variable on the MDS map by applying the partition method of Shin et al. (Korean Journal of Applied Statistics, 28, 1171-1180, 2015). Therefore, we can find information on the similarity among objects as well as find associations among categorical variables using the proposed MDS map.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.2
• /
• pp.227-232
• /
• 2016

3D body scan data are used widely in various fields to make products and living spaces for superior human body fitness. Based on the 3D measurements of human bodies for teens in Size Korea 2013, this research provides a way of finding the representative body sizes and 3D body scan data. First, a multi-dimensional vector space consisting of many measurement items was projected onto a 2D vector space with circumference and length components via factor analysis. The representative body sizes and 3D scan data close to these values were obtained via the Mahalanobis distance in 2D space. Considering the adolescent growth pattern shown on this 2D space, males were divided into 4 age groups and females were divided into 3 age groups. Using the eigenbodies corresponding to the column vectors of the component score coefficient matrix, the representative body sizes of 13 measurement items (male) and 14 measurement items (female) for each age group were calculated. The representative body sizes and 3D scan data are very useful for modeling representative 3D human figures.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.4
• /
• pp.242-249
• /
• 2011

In this paper, the correspondence matching method of stereo images was proposed by means of sampling projective transformation matrix in planar region of scene. Though this study is based on RANSAC, it does not use uniform distribution by random sampling in RANSAC, but use multi non-uniform computed from difference in positions of feature point of image or templates matching. The existing matching method sampled that the correspondence is presumed to correct by use of the condition which the correct correspondence is almost satisfying, and applied RANSAC by matching the correspondence into one to one, but by sampling in stages in multi probability distribution computed for image in the proposed method, the correct correspondence of high probability can be sampled among multi correspondence candidates effectively. In the result, we could obtain many correct correspondence and verify effectiveness of the proposed method in the simulation and experiment of real images.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.18 no.2 s.117
• /
• pp.191-198
• /
• 2007

The electromagnetic coupling between transmission lines in PCB design can degrade the performance of equipment operations. The coupling phenomenon is caused by electromagnetic fields generated by the currents on the transmission lines and the risers. In this paper, an improved method of crosstalk analysis for bent coupled lines on a PCB is proposed and investigated. In the previous cascading method combined with circuit-concept approach, bent coupled lines are devided into sections and each section is represented by ABCD matrix and then they are cascaded. In the proposed method, the crosstalk of bent coupled lines is calculated by the modified circuit-concept approach, where the coupled region is not restricted to the region projected by a generator line on a receptor line but is the total length of receptor line in calculating the forcing terms. Finally, the accuracy of the proposed approach is verified by comparing the calculated results with the measured ones for several bent coupled-line examples.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.4
• /
• pp.23-35
• /
• 2012

This paper presents a 3-D pose (position and orientation) estimation method for an elliptic object in 3-D space. It is difficult to resolve the problem of determining 3-D pose parameters with respect to an elliptic feature in 3-D space by interpretation of its projected feature onto an image plane. As an alternative, we propose a two points-based pose estimation algorithm to seek the 3-D information of an elliptic feature. The proposed algorithm determines a homogeneous transformation uniquely for a given correspondence set of an ellipse and two coplanar points that are defined on model and image plane, respectively. For each plane, two triangular features are extracted from an ellipse and two points based on the polarity in 2-D projection space. A planar homography is first estimated by the triangular feature correspondences, then decomposed into 3-D pose parameters. The proposed method is evaluated through a series of experiments for analyzing the errors of 3-D pose estimation and the sensitivity with respect to point locations.

• Journal of Broadcast Engineering
• /
• v.19 no.6
• /
• pp.866-876
• /
• 2014

Generating of digital hologram of video contents with computer graphics(CG) requires natural fusion of 3D information between real and virtual. In this paper, we propose the system which can fuse real-virtual 3D information naturally and fast generate the digital hologram of fused results using multiple-GPUs based computer-generated-hologram(CGH) computing part. The system calculates camera projection matrix of RGB-Depth camera, and estimates the 3D information of virtual object. The 3D information of virtual object from projection matrix and real space are transmitted to Z buffer, which can fuse the 3D information, naturally. The fused result in Z buffer is transmitted to multiple-GPUs based CGH computing part. In this part, the digital hologram of fused result can be calculated fast. In experiment, the 3D information of virtual object from proposed system has the mean relative error(MRE) about 0.5138% in relation to real 3D information. In other words, it has the about 99% high-accuracy. In addition, we verify that proposed system can fast generate the digital hologram of fused result by using multiple GPUs based CGH calculation.