• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.12
• /
• pp.2681-2690
• /
• 2010

This paper proposes 3D facial modeling system without using 3D scanner and camera or expensive software. This system enables efficient 3D facial modeling to cost reduction and effort saving for natural facial modeling. It detects edges of component of face using edge detection based on fuzzy logic from any 2D image of front face. It was mapped fitting position with 3D standard face model by detected edge more correctly. Also this system generates 3D face model more easily through floating and flexible control and texture mapping after fitting that connection of control point on detected edge from 2D image and mesh of 3D standard face model.

• Tunnel and Underground Space
• /
• v.33 no.6
• /
• pp.508-518
• /
• 2023

This paper presents a new approach for the automatic mapping of discontinuities in a tunnel face based on its 3D digital model reconstructed by LiDAR scan or photogrammetry techniques. The main idea revolves around the identification of discontinuity areas in the 3D digital model of a tunnel face by segmenting its 2D projected images using a deep-learning semantic segmentation model called U-Net. The proposed deep learning model integrates various features including the projected RGB image, depth map image, and local surface properties-based images i.e., normal vector and curvature images to effectively segment areas of discontinuity in the images. Subsequently, the segmentation results are projected back onto the 3D model using depth maps and projection matrices to obtain an accurate representation of the location and extent of discontinuities within the 3D space. The performance of the segmentation model is evaluated by comparing the segmented results with their corresponding ground truths, which demonstrates the high accuracy of segmentation results with the intersection-over-union metric of approximately 0.8. Despite still being limited in training data, this method exhibits promising potential to address the limitations of conventional approaches, which only rely on normal vectors and unsupervised machine learning algorithms for grouping points in the 3D model into distinct sets of discontinuities.

• The Journal of the Korea Contents Association
• /
• v.14 no.11
• /
• pp.620-627
• /
• 2014

In the contemporary age, social environment has changed from analog to digital. As a media is developed, the public is more familiar to digital than analog. In this society, media art rises as new art. One of the most useful methods in the media art is projection mapping. In this paper, I propose a realization process of artwork 'Cloned Me' that is realized by projection mapping method. Artwork is formed by projecting image contents that are visualized a music on a steel structure. It is different from media facade that the object projected contents has a story composition. It is a strength that the object and contents are connected by story and composition. Also it can be modified each other easily for harmony because the structure is assembly-type. A face that is projected by beam projection is consist of unique structure not fixed rectangular frame. Artwork 'Cloned Me' has significance that it reflects the present age to develop digital cloning.

• Journal of the Korean Applied Science and Technology
• /
• v.36 no.4
• /
• pp.1219-1223
• /
• 2019

The skin condition varies by the parts of the face. This study intended to visualize the facial skin based on the characteristics including moisture, sebum, dryness, amount of dead skin cells, and roughness by sectionalizing the facial skin into 9 areas. A total of 98 female in their 20~35s were selected, and the characteristics of skin by area were selected and listed in the order, and the area they felt each characteristic most strongly was analyzed with a weight and the result was visualized. The result showed that the area where they felt dryness the most was around the lips and eyes, and the area where the largest amount of dead skin cells and rough skin existed was the chin. There was a significant increase in moisture content (p <.05), and 90.5% of satisfaction with dryness was satisfied. The result of this study will be used as a basic data for the development of a cosmetic mask pack or a functional cosmetic product for each part.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2004.03b
• /
• pp.946-951
• /
• 2004

ITIS is applied to the several tunnel construction sites in Korea. Tunnel construction properties which are acquired from these sites are transferred to information management server(SQL 2000 server)by client application program in real time. Access permission to DB server depends on the user's roles. Some functions which cannot be embodied in SQL Server are serviced through XML and GMS server is used for spatial data based on GIS part. This system is supposed to give engineers the advantages which are not only easy handling of the program and computerized documentation on every information during construction but also analyzing the acquired data in order to predict the structure of ground and rock mass to be excavated later and show the guideline of construction. Neung-Dong tunnel and Mu-Gua express way tunnel are now under construction and with this system they have 3D visualized map of the geology and tunnel geometry and accumulate database of construction information such as tunnel face mapping results, special notes and pictures of construction and 3D monitoring data, all matters on the stability of rock bolts and shotcrete, and so on. Ground settlement prediction program included in ITIS, based on the artificial neural network(ANN) and supported by GIS technology is applying to the subway tunnel. This prediction tool can make it possible to visualize the ground settlement according to the excavation procedures by contouring the calculated result on 3D GIS map and to assess the damage of buildings in the vicinity of construction site caused by ground settlement.

• Journal of Korean Society for Geospatial Information Science
• /
• v.13 no.2 s.32
• /
• pp.45-50
• /
• 2005

Recently, slope failures are disastrous when they occur in mountainous area adjoining highways. The accidents associated with Slope failures have increased due to rapid urbanization of mountainous area. Therefore, the inspection of slope is conducted to maintain road safety as well as road function. In this study, we apply to the remedy which is comparing existent description to advanced technology using GIS. we utilize a 3D scanner, one of the advanced method, to generate precise and complete road slope model from expert point of view. In result, we are transferred practical data from external slope stability to hazard slope information. We suggest not only the database but also the method of road risk evaluation based on GIS.

• The Journal of Engineering Geology
• /
• v.17 no.4
• /
• pp.555-566
• /
• 2007

In general, RMR classification system is used for the support design of a tunnel. Face mapping during excavation and RMR-based rock classifications are conducted in order to provide information for complementary changes to preliminary survey plans and for continuous geological estimations in direction of tunnel route. Although they are ever so important, there are not enough time for survey in general and sometimes even face mapping is not available. Linear regression analysis for the estimation of mediating RQD and condition of discontinuities, which require longer time and more detailed observation in RMR, was performed and optimum regression equations are suggest as the result. The geological data collected from tunnels were analyzed in accordance with three rock types as sedimentary rock, phyllite and granite to see geological effects, generally not been considered in previous researches. Parameters for the regression analysis were set another RMR factor.

• Tunnel and Underground Space
• /
• v.15 no.3 s.56
• /
• pp.177-184
• /
• 2005

The Sapaesan tunnel, the longest twin tube tunnel (4km) in Korea with 4 lanes each, is under construction with two years of delayed schedule because of the strong opposition from environmental bodies. Therefore, maximizing the construction efficiency was needed in tunnel project to compensate for time delay. This study includes improvements in the construction of the Sapaesan tunnel such as increasing excavation length and changing excavation sequence. In this paper the system for predicting tunnel face ahead is also introduced. Bulk-Emulsion explosive and Cylinder-Cut method were adopted in tunnel blasting to increase the excavation length. Optimum tunnel excavation step was designed to make up delayed time. Tunnel foe mapping, TSP survey and geological prediction system using computerized jumbo-drill were performed fnr safe construction of long and large twin tube tunnel.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2991-3007
• /
• 2022

Two dimensional locality preserving projections (2D-LPP) is an improved algorithm of 2D image to solve the small sample size (SSS) problems which locality preserving projections (LPP) meets. It's able to find the low dimension manifold mapping that not only preserves local information but also detects manifold embedded in original data spaces. However, 2D-LPP is simple and elegant. So, inspired by the comparison experiments between two dimensional linear discriminant analysis (2D-LDA) and linear discriminant analysis (LDA) which indicated that matrix based methods don't always perform better even when training samples are limited, we surmise 2D-LPP may meet the same limitation as 2D-LDA and propose a novel matrix exponential method to enhance the performance of 2D-LPP. 2D-MELPP is equivalent to employing distance diffusion mapping to transform original images into a new space, and margins between labels are broadened, which is beneficial for solving classification problems. Nonetheless, the computational time complexity of 2D-MELPP is extremely high. In this paper, we replace some of matrix multiplications with multiple multiplications to save the memory cost and provide an efficient way for solving 2D-MELPP. We test it on public databases: random 3D data set, ORL, AR face database and Polyu Palmprint database and compare it with other 2D methods like 2D-LDA, 2D-LPP and 1D methods like LPP and exponential locality preserving projections (ELPP), finding it outperforms than others in recognition accuracy. We also compare different dimensions of projection vector and record the cost time on the ORL, AR face database and Polyu Palmprint database. The experiment results above proves that our advanced algorithm has a better performance on 3 independent public databases.

• Science of Emotion and Sensibility
• /
• v.25 no.4
• /
• pp.77-94
• /
• 2022

Social interactions often involve encountering inconsistent information about social others. We conducted a functional magnetic resonance imaging (fMRI) study to comprehensively investigate voxel-wise temporal dynamics showing how impressions are anchored and/or adjusted in response to inconsistent social information. The participants performed a social impression task inside an fMRI scanner in which they were shown a male face, together with a series of four adjectives that described the depicted person's personality traits, successively presented beneath the image of the face. Participants were asked to rate their impressions of the person at the end of each trial on a scale of 1 to 8 (where 1 is most negative and 8 is most positive). We established two hypothetical models that represented two temporal patterns of voxel activity: Model 1 featured decreasing patterns of activity towards the end of each trial, anchoring impressions to initially presented information, and Model 2 showed increasing patterns of activity toward the end of each trial, where impressions were being adjusted using new and inconsistent information. Our data-driven model fitting analyses showed that the temporal activity patterns of voxels within the ventral anterior cingulate cortex, medial orbitofrontal cortex, posterior cingulate cortex, amygdala, and fusiform gyrus fit Model 1 (i.e., they were more involved in anchoring first impressions) better than they did Model 2 (i.e., showing impression adjustment). Conversely, voxel-wise neural activity within dorsal ACC and lateral OFC fit Model 2 better than it did Model 1, as it was more likely to be involved in processing new, inconsistent information and adjusting impressions in response. Our novel approach to model fitting analysis replicated previous impression-related neuroscientific findings, furthering the understanding of neural and temporal dynamics of impression processing, particularly with reference to functionally segmenting each region of interest based on relative involvement in impression anchoring as opposed to adjustment.