• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.3_4
• /
• pp.168-185
• /
• 2003

This paper presents a progressive algorithm that not only can narrow down the search domain in the course of face identification but also can fast reconstruct various 3D objects from a sketch drawing. The sketch drawing, edge-vertex graph without hidden line removal, which serves as input for reconstruction process, is obtained from an inaccurate freehand sketch of a 3D wireframe object. The algorithm is executed in two stages. In the face identification stage, we generate and classify potential faces into implausible, basis, and minimal faces by using geometrical and topological constraints to reduce search space. The proposed algorithm searches the space of minimal faces only to identify actual faces of an object fast. In the object reconstruction stage, we progressively calculate a 3D structure by optimizing the coordinates of vertices of an object according to the sketch order of faces. The progressive method reconstructs the most plausible 3D object quickly by applying 3D constraints that are derived from the relationship between the object and the sketch drawing in the optimization process. Furthermore, it allows the designer to change viewpoint during sketching. The progressive reconstruction algorithm is discussed, and examples from a working implementation are given.

• Archives of Plastic Surgery
• /
• v.35 no.5
• /
• pp.514-520
• /
• 2008

Purpose: Many attempts have been made to describe ideal facial proportions for over two thousands year and constantly lasted till now. Dr. Marquardt has derived supposedly ideal facial proportions from the facial golden mask using golden ratio of 1 : 1.618. On the other hand, facial reducing surgeries such as mandible angle reduction are popularized in Asia because the width of mid and lower face of Korean is recognized to be wider. The purpose is to analyze characters of Korean university students' faces in horizontal plane and establish the objective data for facial width distributions and clinical applications. Methods: We applied the facial golden mask to the photographs in 1000 cases, compared the width of mid and lower face between the facial golden mask and Korean university students' faces. And we first calculated the horizontal ratio(HR) of middle and lower face each for using comparative scale of width, facial golden mask. We divided 1,000 cases into 3 groups by degrees of HR and analyzed data of HR on each groups. Using calculated horizontal ratio, we newly invented the cumulative frequency of distribution graphs in Korean university students' faces. Results: Mean data of HR were over 1.0 in all groups, which means that Korean university students' faces are typically wider than facial golden mask in horizontal planes. And this study was statistically significant(p- value < 0.05). Clinically using the cumulative frequency distributions of Korean university students' face width, we can easily explain changes of facial width to patient after facial reducing surgery and describe the changes into objective data. Conclusion: This study concludes thatKorean university students' faces are wider than facial golden mask is significantly true and the cumulative frequency of distribution graphs are expected to be widely used for comparison of results in facial reducing surgery.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.15 no.4
• /
• pp.101-107
• /
• 2007

This study investigated the wear of the valve and seat insert seating faces. A tester, an exhaust valve and a seat insert were used. Test variables were cycle numbers ($2{\times}10^6,\;4{\times}10^6,\;6{\times}10^6\;and\;8{\times}10^6$) and Hz (10Hz and 25Hz). The other test conditions such as temperature ($350^{\circ}C$), fuel (LPG) and load (1960N) were fixed. The 10Hz tests indicated that the average Rmax of the valve increased at the rate of $7.76{\mu}m/10^6$ cycles starting from $29.42{\mu}m$ at the $2{\times}10^6$ cycles and that of the seat insert increased at the rate of $8.57{\mu}m/10^6$ cycles starting from $34.19{\mu}m$ at the $2{\times}10^6$ cycles. The 25Hz tests indicated that the average Rmax of the valve increased at the rate of $1.58{\mu}m/10^6$ cycles starting from $74.2{\mu}m$ at the $2{\times}10^6$ cycles and that of the seat insert increased at the rate of $1.25{\mu}m/10^6$ cycles starting from $83.95{\mu}m$ at the $2{\times}10^6$ cycles. The tribochemical reaction product covered the two seating faces, preventing the wear of the seating faces. As cycle numbers became greater, the average Rmax of the seating faces became greater, but the increase rate varied significantly depending on the Hz. The wear mechanism of the two faces was investigated through the tribochemical reaction.

• Journal of Digital Contents Society
• /
• v.12 no.1
• /
• pp.1-9
• /
• 2011

JPEG2000 offers the technique which compresses the interested regions with higher quality than the background. It is called by an ROI(Region-of-Interest) coding method. In this paper, we use images including the human faces, which are processed uppermost and compressed with high quality. The proposed method consists of 2 steps. The first step extracts some faces and the second one is ROI coding. To extract the faces, the method cuts or scale-downs some regions with $20{\times}20$ window pixels for all the pixels of the image, and after preprocessing, recognizes the faces using neural networks. Each extracted region is identified by ROI mask and then ROI-coded using Maxshift method. After then, the image is compressed and saved using EBCOT. The existing methods searched the ROI by edge distributions. On the contrary, the proposed method uses human intellect. And the experiment shows that the method is sufficiently useful with images having several human faces.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.6
• /
• pp.75-82
• /
• 2005

The robust feature vector selection method for the multiple frontal face detection is proposed in this paper. The proposed feature vector for the training and classification are integrated by means, amplitude projections, and its 1D Harr wavelet of the input image. And the statistical modeling is performed both for face and nonface classes. Finally, the estimated probability density functions (PDFs) are applied for the detection of multiple frontal faces in the still image. The proposed method can handle multiple faces, partially occluded faces, and slightly posed-angle faces. And also the proposed method is very effective for low quality face images. Experimental results show that detection rate of the propose method is $98.3\%$ with three false detections on the testing data, SET3 which have 227 faces in 80 images.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.7
• /
• pp.896-902
• /
• 2021

Rotated face detection is required in many applications but still remains as a challenging task, due to the large variations of face appearances. In this paper, a polar coordinate transform that is not affected by rotation is proposed. In addition, a method for effectively detecting rotated faces using the transformed image has been proposed. The proposed polar coordinate transform maintains spatial information between facial components such as eyes, mouth, etc., since the positions of facial components are always maintained regardless of rotation angle, thereby eliminating rotation effects. Polar coordinate transformed images are trained using AdaBoost, which is used for frontal face detection, and rotated faces are detected. We validate the detected faces using LBP that trained the non-face images. Experiments on 3600 face images obtained by rotating images in the BioID database show a rotating face detection rate of 96.17%. Furthermore, we accurately detected rotated faces in images with a background containing multiple rotated faces.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.13 no.1
• /
• pp.1-4
• /
• 2003

In the present work we consider morphological structure of the faces of BGO crystals grown by Czochralski technique under the conditions of low temperature gradient (0.1~1 deg/cm) and interconnection between the morphological features of faces at the crystallization front and the formation of defects within the crystal volume. It is demonstrated that the {112} faces retain stability while the growing surface deviates from the crystallographic (112) plane up to 1 degree. At larger deviation, the region of the stable facet growth passes either to the region of macrosteps or to the region of normal growth. depending on conditions.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.1 no.1
• /
• pp.60-65
• /
• 1991

The hydrothermal growth of $GaPO_{4}$ Single Crystals was carried out by the horizontal temperature gradient method. The most promising solvents for the crystal growth of $GaPO_{4}$ are $H_{3}PO_{4}$ and HCl solutions. Single crystals have been hydothermally grown at temperatures over the range $210-290^{\circ}C$ in these solutions with seed crystals. The glowth rates in HCl solution were higher than that for comparable conditions in $H_{3}PO_{4}$ solution. Morphologies of crystals grown at temperatures below $200^{\circ}C$ tended to be bounded by small major rhombohedral(10$\bar{1}$1) faces. In the temperature range from 200 to $430^{\circ}C$, the single crystals have morphologies bounded by prism (10$\bar{1}$0), small major rhombohedral(10$\bar{1}$1) and minor rhombohedral(01$\bar{1}$1) faces at the early stage, and grew with well developed basal(0001) faces by increasing the growth temperature.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.07a
• /
• pp.219-220
• /
• 2016

본 논문에서는 컴퓨터비전 기술 기반의 라이브러리를 이용해 미아 얼굴 정보를 중심으로 매칭을 하는 시스템으로서, 미아 데이터베이스에 등록된 얼굴과 유사한 미아를 정확도 순으로 배열해 주는 시스템을 개발한다. 이는 기존의 텍스트 정보 중심의 미아에 대한 정보 등록 및 조회를 하게 되었을 때 발생하는 정보의 부정확성 등의 문제점을 해결하고 편하고 빠르고 정확하게 정보 입력과 매칭을 함으로써 골든 타임 안에 미아를 찾을 수 있는 장점이 있다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.632-635
• /
• 2003

This paper describes a system fur tracking multiple faces in an input video sequence using facial convex hull based facial segmentation and robust hausdorff distance. The algorithm adapts skin color reference map in YCbCr color space and hair color reference map in RGB color space for classifying face region. Then, we obtain an initial face model with preprocessing and convex hull. For tracking, this algorithm computes displacement of the point set between frames using a robust hausdorff distance and the best possible displacement is selected. Finally, the initial face model is updated using the displacement. We provide an example to illustrate the proposed tracking algorithm, which efficiently tracks rotating and zooming faces as well as existing multiple faces in video sequences obtained from CCD camera.