• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.5
• /
• pp.783-788
• /
• 2002

Recently, research and developmental direction of communication system is concurrent adopting voice data and face image in speaking to provide more higher recognition rate then in the case of only voice data. Therefore, we present a method of lipreading in speech image sequence by using the 3-D facial shape model. The method use a feature information of the face image such as the opening-level of lip, the movement of jaw, and the projection height of lip. At first, we adjust the 3-D face model to speeching face Image sequence. Then, to get a feature information we compute variance quantity from adjusted 3-D shape model of image sequence and use the variance quality of the adjusted 3-D model as recognition parameters. We use the intensity inclination values which obtaining from the variance in 3-D feature points as the separation of recognition units from the sequential image. After then, we use discrete HMM algorithm at recognition process, depending on multiple observation sequence which considers the variance of 3-D feature point fully. As a result of recognition experiment with the 8 Korean vowels and 2 Korean consonants, we have about 80% of recognition rate for the plosives md vowels.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1998.11a
• /
• pp.181-185
• /
• 1998

Recently, research and developmental direction of communication system is concurrent adopting voice data and face image in speaking to provide more higher recognition rate then in the case of only voice data. Therefore, we present a method of lipreading in speech image sequence by using the 3-D facial shape model. The method use a feature information of the face image such as the opening-level of lip, the movement of jaw, and the projection height of lip. At first, we adjust the 3-D face model to speeching face image sequence. Then, to get a feature information we compute variance quantity from adjusted 3-D shape model of image sequence and use the variance quality of the adjusted 3-D model as recognition parameters. We use the intensity inclination values which obtaining from the variance in 3-D feature points as the separation of recognition units from the sequential image. After then, we use discrete HMM algorithm at recognition process, depending on multiple observation sequence which considers the variance of 3-D feature point fully. As a result of recognition experiment with the 8 Korean vowels and 2 Korean consonants, we have about 80% of recognition rate for the plosives and vowels.

• Journal of Sasang Constitutional Medicine
• /
• v.19 no.2
• /
• pp.22-29
• /
• 2007

1. Objectives Sasang Contitutional Medicine, a part of the traditional Korean medical lore, treats illness through a constitutional typing system that categorizespeople into four constitutional types. A few of the important criteria for differentiating the constitutional types are external appearances, inner state of mind, and pathological patterns. We had been developing a 3D Automatic Face Recognition Apparatus (3D-AFRA) in order to evaluate the external appearances with more objectivity. This apparatus provides a 3D image and numerical data on facial configuration, and this study aims to evaluate the mechanical accuracy of the 3D-AFRA hardware. 2. Methods Several objects of different shapes (cube, cylinder, cone, pyramid) were each scanned 10 times using the 3D Automatic Face Recognition Apparatus (3D-AFRA). The results were then compared and analyzed with data retrieved through a laser scanner known for its high accuracy. The error rates were analyzed for each grid point of facial contour scanned with Rapidform2006 (Rapidform2006 is a 3D scanning software that collects grid point data for contours of various products and products and product parts through 3D scanners and other 3D measuring devices; the grid point data thusly acquired is then used to reconstruct highly precise polygon and curvature models). 3. Results and Conclusions The average error rate was 0.22mm for the cube, 0.22mm for the cylinder, 0.125mm for the cone, and 0.172mm for the pyramid. The visual data comparing error rates for measurement figures retrieved with Rapidform2006 is shown in $Fig.3{\sim}Fig.6$. Blue tendency indicates smaller error rates, while red indicates greater error rates The protruding corners of the cube display red, indicating greater error rates. The cylinder shows greater error rates on the edges. The pyramid displays greater error rates on the base surface and around the vertex. The cone also shows greater error around the protruding edge.

• The Journal of the Korea Contents Association
• /
• v.8 no.7
• /
• pp.85-92
• /
• 2008

Meshes are the most appropriate data structures for representing 3D geometries. Surface meshes have been frequently used for representing 3D geometries, which only samples data on the surfaces of the given 3D geometries. Thanks to the improvements of computing powers, it is required to develop more complicated contents which utilize the volumetric information of 3D geometries. In this paper, we introduce a novel volumetric mesh libraries based on the half-face data structure, called OpenVolMesh, and describe its designs and implementations. The OpenVolMesh extends the OpenMesh, which is one of the most famous mesh libraries, by supporting volumetric meshes. The OpenVolMesh provides the generic programming, dynamic allocations of primitive properties, efficient array-based data structures, and source-level compatibility with OpenMesh. We show the usefulness of the OpenVolMesh in the developments of 3D volumetric contents with prototypic implementations such as volumetric mesh smoothing and CW-cell decompositions.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.27-34
• /
• 2023

This paper studied the 3D modeling process for the restoration of the 'Three-story Stone Pagoda of Bulguksa Temple in Gyeongju', a stone pagoda from the Unified Silla Period, using artificial intelligence (AI). Existing 3D modeling methods generate numerous verts and faces, which takes a considerable amount of time for AI learning. Accordingly, a method of performing more efficient 3D modeling by lowering the number of verts and faces is required. To this end, in this study, the structure of the stone pagoda was deeply analyzed and a modeling method optimized for AI learning was studied. In addition, it is meaningful to propose a new 3D modeling methodology for the restoration of stone pagodas in Korea and to secure a data set necessary for artificial intelligence learning.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.11
• /
• pp.465-472
• /
• 2022

In this paper, the style synthesis network is trained to generate style-synthesized video through the style synthesis through training Stylegan and the video synthesis network for video synthesis. In order to improve the point that the gaze or expression does not transfer stably, 3D face restoration technology is applied to control important features such as the pose, gaze, and expression of the head using 3D face information. In addition, by training the discriminators for the dynamics, mouth shape, image, and gaze of the Head2head network, it is possible to create a stable style synthesis video that maintains more probabilities and consistency. Using the FaceForensic dataset and the MetFace dataset, it was confirmed that the performance was increased by converting one video into another video while maintaining the consistent movement of the target face, and generating natural data through video synthesis using 3D face information from the source video's face.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.513-515
• /
• 2019

Techniques to recognize depth values using Kinect infrared projectors continue to evolve. Techniques to track human movements are being developed from the Marcris method to the Bimarris method. Capture of facial movement using Kinect has disadvantages that are not sophisticated. In addition, a method to control the gestures and movements on the face in real time requires much research. Therefore, this paper proposes a technique to create natural 3D image contents by studying technology to apply and control branding technology to extracted face recognition data using Kinect infrared method.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.42 no.6
• /
• pp.988-1001
• /
• 2018

The head and face are the most complicated structures of the human body. However, unlike apparel products, many of the head and face related products are produced in unisex or one-size-fits-all. Therefore, it is necessary to provide a practical analysis that reflects these size structures. This study analyzed head and face measurements of men, women, and both men and women; its purpose lying in clarifying the differences and commonalities between the measurements of men and women and discovering the possibility of integrating analysis on measurements of men and women. 3D face and head measurement data of men and women from 20 to 29 years-old from Size Korea's 6th human body size data were analyzed for this study. The mutual 9 factors were extracted as the results of the three factor analyses (both men and women) separately and men and women integrated. However in the cluster analysis based on the 9 factors, men and women showed mutual characteristics and different landscape in group types and distribution. The commonalities and differences between men and women observed in this study should be applied to the sizes of head and face products.

• Journal of Multimedia Information System
• /
• v.6 no.2
• /
• pp.91-98
• /
• 2019

Many researchers have attempted to recognize three-dimensional faces using feature points extracted from two-dimensional facial photographs. However, due to the limit of flat photographs, it is very difficult to recognize faces rotated more than 15 degrees from original feature points extracted from the photographs. As such, it is difficult to create an algorithm to recognize faces in multiple angles. In this paper, it is proposed a new algorithm to recognize three-dimensional face recognition based on feature points extracted from a flat photograph. This method divides into six feature point vector zones on the face. Then, the vector value is compressed and expanded according to the rotation angle of the face to recognize the feature points of the face in a three-dimensional form. For this purpose, the average of the compressibility and the expansion rate of the face data of 100 persons by angle and face zone were obtained, and the face angle was estimated by calculating the distance between the middle of the forehead and the tail of the eye. As a result, very improved recognition performance was obtained at 30 degrees of rotated face angle.

• Journal of International Society for Simulation Surgery
• /
• v.1 no.2
• /
• pp.57-61
• /
• 2014

We present a robust 3D facial reconstruction method using a single image generated by face-specific super resolution technique. Based on the several consecutive frames with low resolution, we generate a single high resolution image and a three dimensional facial model based on it. To do this, we apply PME method to compute patch similarities for SR after two-phase warping according to facial attributes. Based on the SRI, we extract facial features automatically and reconstruct 3D facial model with basis which selected adaptively according to facial statistical data less than a few seconds. Thereby, we can provide the facial image of various points of view which cannot be given by a single point of view of a camera.