• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.12
• /
• pp.509-516
• /
• 2022

Accurate pose prediction of objects in 3D space is an important visual recognition technique widely used in many applications such as scene understanding in both indoor and outdoor environments, robotic object manipulation, autonomous driving, and augmented reality. Most previous works for object pose estimation have the limitation that they require an exact 3D CAD model for each object. Unlike such previous works, this paper proposes a novel neural network model that can predict the poses of unknown objects based on only their RGB color images without the corresponding 3D CAD models. The proposed model can obtain depth maps required for unknown object pose prediction by using an adaptive depth estimator, AdaBins,. In this paper, we evaluate the usefulness and the performance of the proposed model through experiments using benchmark datasets.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.25 no.2
• /
• pp.117-131
• /
• 2007

Representation of building internal space is an active research area as the need for more geometrically accurate and visually realistic increases. 3 dimensional representation is common ground of research for disciplines such as computer graphics, architectural design and engineering and Geographic Information System (GIS). In many cases CAD plans are the starting point of reconstruction of 3D building models. The main objectives of building reconstruction in GIS applications are visualization and spatial analysis. Hence, CAD plans need to be preprocessed and edited to adapt to the data models of GIS SW and then georeferenced to enable spatial analysis. This study automated the preprocessing of CAD data using AutoCAD VBA (Visual Basic Application), and the processed data was topologically restructured for further analysis in GIS environment. Accuracy of georeferencing CAD data was also examined by comparing the results of coordinate transformation by using digital maps and GPS measurements as the sources of ground control points. The reconstructed buildings were then applied to visualization and network modeling.

• Journal of KIBIM
• /
• v.10 no.2
• /
• pp.1-11
• /
• 2020

BIM(Building Information Modeling) is a salient technology for influential innovation in the construction industry. The patent network analysis is useful for suggesting the direction of technology development and exploring the research and development field. Therefore, the purpose of this study is to analyze the BIM technology structure and core technologies according to the convergence of BIM technology and market expansion. In this study, social network analysis was conducted by establishing a co-classification IPC network for the United States BIM patent. In particular, the characteristics of the major technical areas in the BIM technology network were identified through centrality analysis. G06F017/00, digital computing or data processing method, is a core technology field in the BIM network. Arrangements, apparatus or systems for transmission of digital information, H04L029/00 is an influential technology across the network. B25J009/00 for program controlled manipulators is an intermediary technology field and G06T019/00, manipulating 3D models or images for computer graphics, is an important field for technological development competitiveness.

• Journal of the Korean Institute of Telematics and Electronics T
• /
• v.35T no.1
• /
• pp.125-132
• /
• 1998

In this paper, radio-wave propagation characteristics for L-band frequency are investigated using real data in case of base station in universal domestic environment. L-band frequency was recently emphasized the importance in PCS network. Measure+ments are carried out on vertical polarization characteristics in the 1.805 GHz frequency, three different estimated models, 3km 300m and 30m, are measured separately for 3 or 4 months at TX/RX fixed states. The experimental results of the estimated models are indicating the difference of 1~6dB compared to the theoretical value. The results present useful data in determining the base station sites.

• Geomechanics and Engineering
• /
• v.31 no.3
• /
• pp.291-304
• /
• 2022

Maximum shear modulus (G_max or G₀) is an important soil property useful for many engineering applications, such as the analysis of soil-structure interactions, soil stability, liquefaction evaluation, ground deformation and performance of seismic design. In the current study, bender element (BE) tests are used to evaluate the effect of the void ratio, effective confining pressure, grading characteristics (D₅₀, C_u and C_c), anisotropic consolidation and initial fabric anisotropy produced during specimen preparation on the G_max of sand-gravel mixtures. Based on the tests results, an empirical equation is proposed to predict G_max in granular soils, evaluated by the experimental data. The artificial neural network (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS) models were also applied. Coefficient of determination (R²) and Root Mean Square Error (RMSE) between predicted and measured values of G_max were calculated for the empirical equation, ANN and ANFIS. The results indicate that all methods accuracy is high; however, ANFIS achieves the highest accuracy amongst the presented methods.

• Tunnel and Underground Space
• /
• v.22 no.2
• /
• pp.93-105
• /
• 2012

This paper compares the two- and three-dimensional (2D and 3D) approaches for the numerical determination of the equivalent mechanical properties of fractured rock masses. Both orthogonally-fractured model and discrete fracture networks (DFN) were used for the geometry and 2D models were cut in various directions from 3D model to compare their mechanical properties. Geological data were loosely based on the data available from Sellafield, UK. Analytical method based on compliance tensor transformation was used for investigation in orthogonally fractured rock and numerical experiments were conducted on fractured rock mass with DFN geometry. It is shown that 2D approach always overestimates the elastic modulus of fractured rock masses by a factor of up to around two because fractures are assumed to be perpendicular to the model plane in 2D problems. Poisson ratios tend to have larger values in 2D analysis while there is opposite trend in some sections. The study quantitatively demonstrates the limitation of the 2D approach that uses the simplified model from true 3D geometry.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.38 no.2
• /
• pp.19-27
• /
• 2001

All accurate and efficient antenna feed model is very important for computing GPR response using the FDTD method In literature, there are several feed models such as the equivalent network in angular-frequency domain, 1-D transmission-line cell， voltage boundary condition in time domain, etc. In this paper, theoretical relationship among the models is investigated. It is found that the above three models become equivalent when a short and lossless feed line can match with its connected transmitter receiver). In view of accuracy and efficiency of the simulation, the FDTD results according to the feed models arc compared with the measured data of the receiving responses for an actual GPR system.

• Journal of the Korean Society of Industry Convergence
• /
• v.27 no.3
• /
• pp.563-572
• /
• 2024

Image processing and computer vision technologies are becoming increasingly important in a variety of application fields that require techniques and tools for sophisticated image analysis. In particular, image segmentation is a technology that plays an important role in image analysis. In this study, in order to identify recent research trends on image segmentation techniques, we used the Web of Science(WoS) database to analyze the R&D topography based on the network structure of the author's keyword co-occurrence matrix. As a result, from 2015 to 2023, as a result of the analysis of the R&D map of research articles on image segmentation, R&D in this field is largely focused on four areas of research and development: (1) researches on collecting and preprocessing image data to build higher-performance image segmentation models, (2) the researches on image segmentation using statistics-based models or machine learning algorithms, (3) the researches on image segmentation for medical image analysis, and (4) deep learning-based image segmentation-related R&D. The scientometrics-based analysis performed in this study can not only map the trajectory of R&D related to image segmentation, but can also serve as a marker for future exploration in this dynamic field.

• Journal of Digital Contents Society
• /
• v.19 no.3
• /
• pp.435-443
• /
• 2018

Recently, with the explosive development of computing power, various methods such as RNN and CNN have been proposed under the name of Deep Learning, which solve many problems of Computer Vision have. The Generative Adversarial Network, released in 2014, showed that the problem of computer vision can be sufficiently solved in unsupervised learning, and the generation domain can also be studied using learned generators. GAN is being developed in various forms in combination with various models. Machine learning has difficulty in collecting data. If it is too large, it is difficult to refine the effective data set by removing the noise. If it is too small, the small difference becomes too big noise, and learning is not easy. In this paper, we apply a deep CNN model for extracting facial region in image frame to GAN model as a preprocessing filter, and propose a method to produce composite images of various facial expressions by stably learning with limited collection data of two persons.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.792-799
• /
• 2022

Building Information Modeling (BIM) technology is a key component of modern construction engineering and project management workflows. As-is BIM models that represent the spatial reality of a project site can offer crucial information to stakeholders for construction progress monitoring, error checking, and building maintenance purposes. Geometric methods for automatically converting raw scan data into BIM models (Scan-to-BIM) often fail to make use of higher-level semantic information in the data. Whereas, semantic segmentation methods only output labels at the point level without creating object level models that is necessary for BIM. To address these issues, this research proposes a hybrid semantic-geometric approach for clutter-resistant floorplan generation from laser-scanned building point clouds. The input point clouds are first pre-processed by normalizing the coordinate system and removing outliers. Then, a semantic segmentation network based on PointNet++ is used to label each point as ceiling, floor, wall, door, stair, and clutter. The clutter points are removed whereas the wall, door, and stair points are used for 2D floorplan generation. A region-growing segmentation algorithm paired with geometric reasoning rules is applied to group the points together into individual building elements. Finally, a 2-fold Random Sample Consensus (RANSAC) algorithm is applied to parameterize the building elements into 2D lines which are used to create the output floorplan. The proposed method is evaluated using the metrics of precision, recall, Intersection-over-Union (IOU), Betti error, and warping error.