Department News
State-of-the-Art Optical 3D-Measurement and 3D-Data Processing (안성준)
Seminar Date
2002-11-11
Author
박경시
Date
2002-11-11
Views
1615
1. 제 목 : State-of-the-Art Optical 3D-Measurement and 3D-Data Processing
2. 연 사 : 안 성 준 (Dept. of Information Processing, Fraunhofer IPA, Germany)
3. 일 시 : 2002년 11월 28일 (목) 16:00 - 17:00
4. 장 소 : 서울대 신공학관(301동) 1512호 세미나실
5. 내 용 :
Parametric model recovery of real objects is a central subject in coordinate metrology, reverse engineering, and computer/machine vision, with which the objects are to be represented by their model parameters estimated from the measurement data (point cloud). The demand on parametric model recovery is practically everywhere, e.g., from particle physic, drug design, manufacturing engineering, and entertainment industry. And the 3D-data volumes to be processed are exploding as the scientific/technical progresses in each application field continue.
The overall procedures of parametric model recovery can be divided into two parts, the 3D-measurement and the 3D-data processing. This talk deals with the state-of-the-art methods and techniques for optical 3D-measurement and 3D-data processing. First, we review the principles of a variety of optical 3D-measurement techniques including close-range photogrammetry, stereo vision, structured-light method, and laser radar. In particular, the system (camera and/or projector) calibration techniques will be discussed in detail. Next, we talk about the very new algorithmic techniques for 3D-data processing, the Orthogonal Distance Fitting (ODF) estimating the parameters of implicit/parametric curve/surface in 2D/3D-space by minimizing the square sum of the shortest distances between the model feature and the measurement points. A few examples of parametric model recovery will be shown.
6. 약 력 :
85.2 서울대학교 기계설계학과 학사
87.2 KAIST 기계공학과 석사 (생산공학)
94.10 Dipl.-Ing., Construction and Production Engineering
90. ~ 현재 Dept. of Information Processing, Fraunhofer IPA, Germany
7. 문 의 : 기계항공공학부 이교일 교수, 민경덕 교수, 강연준 교수 (☏ 880-1691)
서울대 BK21 기계분야사업단 행정실 (☏ 880-7132)
2. 연 사 : 안 성 준 (Dept. of Information Processing, Fraunhofer IPA, Germany)
3. 일 시 : 2002년 11월 28일 (목) 16:00 - 17:00
4. 장 소 : 서울대 신공학관(301동) 1512호 세미나실
5. 내 용 :
Parametric model recovery of real objects is a central subject in coordinate metrology, reverse engineering, and computer/machine vision, with which the objects are to be represented by their model parameters estimated from the measurement data (point cloud). The demand on parametric model recovery is practically everywhere, e.g., from particle physic, drug design, manufacturing engineering, and entertainment industry. And the 3D-data volumes to be processed are exploding as the scientific/technical progresses in each application field continue.
The overall procedures of parametric model recovery can be divided into two parts, the 3D-measurement and the 3D-data processing. This talk deals with the state-of-the-art methods and techniques for optical 3D-measurement and 3D-data processing. First, we review the principles of a variety of optical 3D-measurement techniques including close-range photogrammetry, stereo vision, structured-light method, and laser radar. In particular, the system (camera and/or projector) calibration techniques will be discussed in detail. Next, we talk about the very new algorithmic techniques for 3D-data processing, the Orthogonal Distance Fitting (ODF) estimating the parameters of implicit/parametric curve/surface in 2D/3D-space by minimizing the square sum of the shortest distances between the model feature and the measurement points. A few examples of parametric model recovery will be shown.
6. 약 력 :
85.2 서울대학교 기계설계학과 학사
87.2 KAIST 기계공학과 석사 (생산공학)
94.10 Dipl.-Ing., Construction and Production Engineering
90. ~ 현재 Dept. of Information Processing, Fraunhofer IPA, Germany
7. 문 의 : 기계항공공학부 이교일 교수, 민경덕 교수, 강연준 교수 (☏ 880-1691)
서울대 BK21 기계분야사업단 행정실 (☏ 880-7132)
