• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.7
• /
• pp.1250-1261
• /
• 2002

Many research works have validated the J-T approach to elastic-plastic crack-tip stress fields in a variety of plane strain specimens. To generalize the validity of J-T method, further investigations are however needed for more practical 3D structures than the idealized plane strain specimens. In this work, we perform 3D finite element (FE) modeling of welded plate and straight pipe, and accompanying elastic, elastic-plastic FE analyses. Manual 3D modeling is almost prohibitive, since the models contain semi-elliptical interfacial cracks which require singular elements. To overcome this kind of barrier, we develop a program generating the meshes for semi-elliptical interfacial cracks. We then compare the detailed 3D FE stress fields to those predicted with J-T two parameters. Thereby we extend the validity of J-T application to 3D structures and infer some useful informations for the design or assessment of pipe welds.

• The Journal of Korean Association of Computer Education
• /
• v.9 no.1
• /
• pp.89-95
• /
• 2006

The local geometric properties such as curvatures and normal vectors play important roles for analyzing the local shape of objects in the fields of computer graphics and computer vision. The result of the geometric operations such as mesh simplification and mesh smoothing is dependent on how to compute the curvatures of meshes because there is no exact mathematical definition of curvature at vertices on 3D meshes. Therefore, In this paper, we indicate the fatal error in computing the sectional curvatures of the most previous discrete curvature estimations. Moreover, we present a discrete curvature estimation to overcome the error, which is based on the parabola interpolation and the geometric properties of Bezier curves. Therefore, We can well distinguish between the sharp vertices and the flat ones, so our method may be applied to a variety of geometric operations.

• Journal of Korea Multimedia Society
• /
• v.9 no.11
• /
• pp.1483-1495
• /
• 2006

The navigation mesh represents a terrain as a set of triangles on which characters may move around. The navigation mesh cab be generated automatically, and it is more flexible in representing 3D surface. The number of triangles to represent a terrain may vary according to the structure of the terrain. As characters are moving around on a navigation mesh, the path planning can be performed more easily by projecting the 3D surfaces into 2D space. However, when the terrain is represented with an elaborated mesh of large number of triangles to achieve more realistic movements, the path finding can be very inefficient because there are too many states(triangles) to be searched. In this paper, we propose an efficient method of path finding in 3D games where the terrain is represented by navigation meshes. Our method uses the visibility tests. When the graph-based search is applied to elaborated polygonal meshes for detailed terrain representation, the path finding can be very inefficient because there are too many states(polygons) to be searched. In our method, we reduce the search space by using visibility tests so that the search can be fast even on the detailed terrain with large number of polygons. First we find the visible vertices of the obstacles, and define the heuristic function as the distance to the goal through those vertices. By doing that, the number of states that the graph-based search visits can be substantially reduced compared to the plane search with straight-line distance heuristic.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.4C
• /
• pp.472-484
• /
• 2004

We proposed 3D mesh watermarking algorithm using CEGI distribution. In the proposed algorithm, we divide a 3D mesh of VRML data into 6 patches using distance measure and embed the same watermark bits into the normal vector direction of meshes that mapped into the cells of each patch that have the large magnitude of complex weight of CEGI. The watermark can be extracted based on the known center point of each patch and order information of cell. In an attacked model by affine transformation, we accomplish the realignment process before the extraction of the watermark. Experiment results exhibited the proposed algorithm is robust by extracting watermark bit for geometrical and topological deformed models.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2002.04a
• /
• pp.105-112
• /
• 2002

A new grid-based approach is presented for automatic generation of hexahedral meshes for simulation of plastic deformation in metal forming. In this approach, special enveloping schemes are applied, to eradicate the sources of the degenerate elements that may appear in a generated mesh. The schemes are described in detail, along with a complete procedure for mesh generation. The capability of the approach to deal with an arbitrary, 3-D process geometry is demonstrated through application to a selected forming problem.

• Journal of the Korea Computer Graphics Society
• /
• v.15 no.3
• /
• pp.19-26
• /
• 2009

In this paper, a new adaptive vertex clustering using a KD-tree is presented for 3D mesh partitioning. A vertex clustering is used to divide a huge 3D mesh into several partitions for various mesh processing. An octree-based clustering and K-means clustering are currently leading techniques. However, the octree-based methods practice uniform space divisions and so each partitioned mesh has non-uniformly distributed number of vertices and the difference in its size. The K-means clustering produces uniformly partitioned meshes but takes much time due to many repetitions and optimizations. Therefore, we propose to use a KD-tree to efficiently partition meshes with uniform number of vertices. The bounding box region of the given mesh is adaptively subdivided according to the number of vertices included and dynamically determined axis. As a result, the partitioned meshes have a property of compactness with uniformly distributed vertices.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.27 no.1
• /
• pp.126-137
• /
• 2003

Many research works have been performed on the J-T approach for elastic-plastic crack-tip stress fields in a variety of plane strain specimens. To generalize the validity of J-T method, further investigations are however needed fur more practical 3D structures than the idealized plane strain specimens. The present study deals mainly with 3D finite element (FE) modeling of welded plate and straight pipe, and accompanying elastic, elastic-plastic FE analyses. Manual 3D modeling is almost prohibitive, since the models contain semi-elliptical interfacial cracks which require singular elements. To overcome this kind of barrier, we develop a program generating the meshes fur semi-elliptical interfacial cracks. We then compare the detailed 3D FE stress fields to those predicted with J-T two parameters. The validity of J-T approach is thereby extended to 3D yield-strength-mismatched weld joints, and useful information is inferred fur the design or assessment of pipe welds.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1999.11b
• /
• pp.47-52
• /
• 1999

The visualization of Moire or laser-scanned data has been explored by many researchers and has been an important issue on computer graphics research. In this paper, we present various techniques that handle tremendous amount of 3D range data which are generated by the ETRI- Moire Scanner. The techniques include constructing an efficient data structure, constructing triangle meshes and decimation and registration of multiple-view range images and textures.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.141-148
• /
• 2005

Simplification of 3D models is becoming necessary with popularity of 3D graphics over mobile or the internet channels with limited channel capacity. Surfaces of a 3D model are usually approximated by a series of triangular meshes, and vertex contraction method is employed widely to minimize the deviation from the original model. Determination of the best position after contraction depends on the calculation of simplification error. We propose a new measure for computing the error so that the simplified model represents the original faithfully. We demonstrate the improved results with real 3D models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.670-684
• /
• 2024

3D Cross-Modal Retrieval (3DCMR) is a task that retrieves 3D objects regardless of modalities, such as images, meshes, and point clouds. One of the most prominent methods used for 3DCMR is the Cross-Modal Center Loss Function (CLF) which applies the conventional center loss strategy for 3D cross-modal search and retrieval. Since CLF is based on center loss, the center features in CLF are also susceptible to subtle changes in hyperparameters and external inferences. For instance, performance degradation is observed when the batch size is too small. Furthermore, the Mean Squared Error (MSE) used in CLF is unable to adapt to changes in batch size and is vulnerable to data variations that occur during actual inference due to the use of simple Euclidean distance between multi-modal features. To address the problems that arise from small batch training, we propose a Noisy Center Loss (NCL) method to estimate the optimal center features. In addition, we apply the simple Siamese representation learning method (SimSiam) during optimal center feature estimation to compare projected features, making the proposed method robust to changes in batch size and variations in data. As a result, the proposed approach demonstrates improved performance in ModelNet40 dataset compared to the conventional methods.