• Journal of the Korea Computer Graphics Society
• /
• v.15 no.1
• /
• pp.7-15
• /
• 2009

In this paper, a novel framework that synthesizes and clones facial expression in parameter spaces is presented. To overcome the difficulties in manipulating face geometry models with high degrees of freedom, many parameterization methods have been introduced. In this paper, a data-driven parameterization method is proposed that represents a variety of expressions with a small set of fundamental independent movements based on the ICA technique. The face deformation due to the parameters is also learned from the data to capture the nonlinearity of facial movements. With this parameterization, one can control the expression of an animated character's face by the parameters. By separating the parameterization and the deformation learning process, we believe that we can adopt this framework for a variety applications including expression synthesis and cloning. The experimental result demonstrates the efficient production of realistic expressions using the proposed method.

• Korean Journal of Culture and Social Issue
• /
• v.20 no.4
• /
• pp.347-364
• /
• 2014

The present study was conducted to investigate how emotional expression change, test delay, and background influence on face recognition. In experiment 1, participants were presented with negative faces at study phase and administered for standard old-new recognition test including targets of negative and neutral expression for the same faces. In experiment 2, participants were studied negative faces and tested by old-new face recognition test with targets of negative and positive faces. In experiment 3, participants were presented with neutral faces at study phase and had to identify the same faces with no regard for negative and neutral expression at face recognition test. In all three experiments, participants were assigned into either immediate test or delay test, and target faces were presented in both white and black background. Results of experiments 1 and 2 indicated higher rates for negative faces than neutral or positive faces. Facial expression consistency enhanced face recognition memory. In experiment 3, the superiority of facial expression consistency were demonstrated by higher rates for neutral faces at recognition test. If facial expressions were consistent across encoding and retrieval, memory performance on face recognition were enhanced in all three experiments. And the effect of facial expression change have different effects on background conditions. The findings suggest that facial expression change make face identification hard, and time and background also affect on face recognition.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.315-318
• /
• 2007

자동 얼굴 인식, 표정 인식과 같은 얼굴 영상과 관련된 다양한 연구 분야는 일반적으로 입력 얼굴 영상에 대한 정규화가 필요하다. 사람의 얼굴은 표정, 조명 등에 따라 다양한 형태변화가 있어 입력 영상 마다 정확한 대표 특징 점을 찾는 것은 어려운 문제이다. 특히 감고 있는 눈이나 작은 눈 등은 검출하기 어렵기 때문에 얼굴 관련 연구에서 성능을 저하시키는 주요한 원인이 되고 있다. 이에 다양한 변화에 강건한 눈 검출을 위하여 본 논문에서는 눈의 텍스처 정보를 이용한 눈 검출 방법을 제안한다. 얼굴 영역에서 눈의 텍스처가 갖는 특성을 정의하고 두 가지 형태의 Eye 필터를 정의하였다. 제안된 방법은 Adaboost 기반의 얼굴 영역 검출 단계, 조명 정규화 단계, Eye 필터를 이용한 눈 후보 영역 검출 단계, 눈 위치 점 검출 단계 등 총 4단계로 구성된다. 실험 결과들은 제안된 방법이 얼굴의 자세, 표정, 조명 상태 등에 강건한 검출 결과를 보여주며 감은 눈 영상에서도 강건한 결과를 보여준다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.4
• /
• pp.287-293
• /
• 2018

It is prerequisite to generate epipolar resampled images by reducing the y-parallax for accurate and efficient processing of satellite stereo images. Minimizing y-parallax requires the accurate sensor modeling that is carried out with ground control points. However, the approach is not feasible over inaccessible areas where control points cannot be easily acquired. For the case, a relative orientation can be utilized only with conjugate points, but its accuracy for satellite sensor should be studied because the sensor has different geometry compared to well-known frame type cameras. Therefore, we carried out the bias-compensation of RPCs (Rational Polynomial Coefficients) without any ground control points to study its precision and effects on the y-parallax in epipolar resampled images. The conjugate points were generated with stereo image matching with outlier removals. RPCs compensation was performed based on the affine and polynomial models. We analyzed the reprojection error of the compensated RPCs and the y-parallax in the resampled images. Experimental result showed one-pixel level of y-parallax for Kompsat-3 stereo data.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.1799-1803
• /
• 2008

표면영상유속계는 하천 표면의 영상을 분석하여 유속을 산정하는 매우 실용적이며 간편한 장비이다. 그러나, 표면영상유속계를 이용하여 유량을 산정하고자 할 경우, 하천 표면의 평면 측량 자료와 하천의 단면 측량 자료가 반드시 필요하다. 이 때문에 표면영상유속계의 간편성과 유용성에도 불구하고, 이용자들이 쉽게 이용하기 어렵다는 그릇된 인식을 줄 수 있다. 만일 효율적이고 간편하게 하천의 단면을 추정할 수 있다면, 표면영상유속계를 마치 일반적인 프로펠러 유속계처럼 쉽게 이용할 수 있을 것이다. 이 연구는 일반적인 평면 측량없이, 두 대의 비디오 카메라로 이루어진 표면 영상 유속계를 이용하여 하천 평면을 계측하는 방법을 개발하는 것이다. 이를 통하여 표면 영상 분석 과정을 반자동화할 수 있게 된다. 두 대의 카메라를 이용한 평면 측량은 사진 측량 분야이나 컴퓨터 비전 분야에서 오랫 동안 연구되어 왔다. 이 기법을 표면영상유속계에 적용함으로써 간단하게 하천의 평면 좌표를 구할 수 있도록 하였다. 두 대의 카메라에 대해서는 직접 선형 변환법을 이용하여 내부 표정과 외부 표정을 수행하여 변환의 매개 변수들을 추정하였다. 추정된 변수들과 공간 전방 교회법을 이용하여 하천의 고정된 기준점들의 좌표를 측정한다. 측정된 좌표점들은 기울어진 영상을 연직으로 사영된 평면으로 변환하는 데 이용되며, 이 과정을 통하여 번거로운 하천의 평면 측량 과정을 생략할 수 있게 되었다. 온천천에 실제 적용하여 본 결과, 결과는 아직은 만족할 만한 정도는 아니나, 보다 정밀한 카메라의 보정 등을 통하여 보다 나은 결과를 도출할 수 있을 것으로 기대된다.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2010.04a
• /
• pp.129-131
• /
• 2010

This study is to measure 3d deformation of railroad-bridge of steel structure using the non-metric high-resolution digital camera. Measuring the deformation is to be utilized relative orientation using the coplanarity for reduction of the field survey and efficiency of the work. The results are compared with deformation by exterior orientation parameters, which are computed from 3d measurement of control points by the Total-station. Then accuracy of the utilized method will be verified.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.4
• /
• pp.305-315
• /
• 2020

The purpose of this study is to compare algorithms of single photo resection, which determines the exterior orientation parameters used in fields such as photogrammetry, computer vision, robotics, etc. To this end, the algorithms were compared by generating experimental data by simulating terrain based on a camera used in aerial and close-range photogrammetry. Through experiments on aerial photographic camera that was taken almost vertically, it was possible to determine the exterior orientation parameters using three ground control points, but the Procrustes algorithm was sensitive to the configuration of the ground control points. Even in experiments with a close-range amateur camera where the attitude angles of the camera change significantly, the algorithm was sensitive to the configuration of the ground control points, and the other algorithms required at least six ground control points. Through experiments with two types of cameras, it was found that cosine lawbased spatial resection shows performance similar to that of a traditional photogrammetry algorithm because the number of iterations is short and no explicit initial values are required.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.14 no.1
• /
• pp.9-16
• /
• 1996

A new stereo matching method for the relative orientation and the photogrammetric triangulation has been pro-posed. It matches the pairs of conjugate points to be used as pass points and tie points in digital aerial photographs instead of selecting these points with manual point transfer and measurements. Three unique steps included in the proposed matching method are as followings. The first step is searching interest points, and designating them as candidate points to be matched. The second is matching the points from the pair of images by the Cross Correlation Method in both direction(left to right direction and right to left). The third is selecting consistent pairs in the both matching directions. Computer programs based on the proposed matching method have been developed, and with digital aerial photographs which have full ground coordinate information tests were performed to know reliabiliy and positional accuracy of proposed method. Results of the tests reveal that the proposed matching method can eliminate the in-correctly matched pairs more efficiently than other matching methods, thus this can be more reliably applied to the relative orientation and the photogrammetric triangulation.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.3 no.4
• /
• pp.276-281
• /
• 2008

Because the distribution systems experience frequently the fault by several causes, the identification task of fault location plays very important role in the view point of power supply reliability. The distribution systems are designed as radial structure with three-phase and single-phase branch line to supply the electric power to the widely dispersed loads, and it have a several load taps within each line segment. it makes the accurate fault distance determination difficult. Accordingly in this papers, the existing fault point determination methods are surveyed and analyzed, and then a fault distance determination method for distribution feeder is adopted which can be executed effectively in DAS center. Finally, the adopted method is verified using EMTP simulation.

• Journal of Korea Spatial Information System Society
• /
• v.5 no.2 s.10
• /
• pp.83-89
• /
• 2003

In this study, a new coded targets that being suitable for a Mobile Mapping System and method of automated target recognition and coordinates determination of center point was developed, the purpose of a Mobile Mapping System is acquisition and logging data around the road for the camera calibration and the Exterior orientation of system, target installed previously and CCD cameras captured the image, target center point was observed by hands, but using coded target, target recognition and center point observation process will be automated effectively, in addition we keeped up the consistency of photo coordinates observation.