• Journal of the Korea Computer Graphics Society
• /
• v.15 no.2
• /
• pp.19-34
• /
• 2009

This paper presents an automatic rendering system for 3D meshes which generates images in the style of oriental black ink painting. In this paper, we analyze the characteristics of black ink painting and present some requirements for our black ink rendering system. Then we survey the steps used by artists to create black ink paintings. Based on the analysis of drawing steps, we propose a black ink rendering system which reproduces the key features of black ink painting through three sub-systems: the feature line layer, the interior shading layer, and the media layer. The main contribution of this paper is to introduce new methods which generate realistic and natural results very fast while any simulation methods are not used to satisfy real-time performance. We describe our implementation of each of these, and demonstrate our results in the major three styles of black ink painting: the outline style, the spontaneous style, and the outline-spontaneous style. Our system renders 3D models of moderate complexity at interactive frame rates. As a result, we expect that our system can be directly applied to real-time applications such as computer games and virtual environments.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.3
• /
• pp.23-29
• /
• 2022

Virtual humans are implemented with dedicated modeling tools like Unity 3D Engine in virtual space (virtual reality, mixed reality, metaverse, etc.). Various human modeling tools have been introduced to implement virtual human-like appearance, voice, expression, and behavior similar to real people, and virtual humans implemented via these tools can communicate with users to some extent. However, most of the virtual humans so far have stayed unimodal using only text or speech. As AI technologies advance, the outdated machine-centered dialogue system is now changing to a human-centered, natural multi-modal system. By using several pre-trained networks, we implemented an emotion-based multi-modal dialogue system, which generates human-like utterances and displays appropriate facial expressions in real-time.

• Journal of the Korea Computer Graphics Society
• /
• v.20 no.2
• /
• pp.1-11
• /
• 2014

The quality of 2D-to-3D conversion depends on the accuracy of the assigned depth to scene objects. Manual depth painting for given objects is labor intensive as each frame is painted. Specifically, a human is one of the most challenging objects for a high-quality conversion, as a human body is an articulated figure and has many degrees of freedom (DOF). In addition, various styles of clothes, accessories, and hair create a very complex silhouette around the 2D human object. We propose an efficient method to estimate visually pleasing depths of a human at every frame in a monocular video. First, a 3D template model is matched to a person in a monocular video with a small number of specified user correspondences. Our pose estimation with sequential joint angular constraints reproduces a various range of human motions (i.e., spine bending) by allowing the utilization of a fully skinned 3D model with a large number of joints and DOFs. The initial depth of the 2D object in the video is assigned from the matched results, and then propagated toward areas where the depth is missing to produce a complete depth map. For the effective handling of the complex silhouettes and appearances, we introduce a partial depth propagation method based on color segmentation to ensure the detail of the results. We compared the result and depth maps painted by experienced artists. The comparison shows that our method produces viable depth maps of humans in monocular videos efficiently.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.1
• /
• pp.27-33
• /
• 2018

This paper presents a new adaptive LSB steganography for high capacity in spatial color images. The number of least signi ficant bit (LSB) of each RGB component in a color image pixel, to replace with the data bits to be hidden, was determine d through analysis of the worst case peak signal noise ratio (PSNR). In addition, the combination of the number of bits is determined adaptively according to image content. That is, 70% of the data to be hidden is proposed to be replaced with 3 bit LSB of two components, 2 bit LSB of the rest component, and 30% be replaced with 4 bit LSB of each RGB compon ent. To find edge areas in an image, delta sorting in local area is also suggested. Using the proposed method, the data cap acity is 9.2 bits per pixel (bpp). The average PSNR value of the tested images with concealed data of up to 60Kbyte was 43.9 db and also natural histograms were generated.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.9
• /
• pp.21-31
• /
• 2014

Singular value decomposition (SVD) has been widely used to identify unique features from a data set in various fields. However, a complex matrix calculation of SVD requires tremendous computation time. This paper improves the performance of a representative one-sided block Jacoby algorithm using a two-dimensional (2D) multi-core system. In addition, this paper explores an optimal multi-core system by varying the number of processing elements in the 2D multi-core system with the same 400MHz clock frequency and TSMC 28nm technology for each matrix-based one-sided block Jacoby algorithm ($128{\times}128$, $64{\times}64$, $32{\times}32$, $16{\times}16$). Moreover, this paper demonstrates the potential of the 2D multi-core system for the one-sided block Jacoby algorithm by comparing the performance of the multi-core system with a commercial high-performance graphics processing unit (GPU).

• Journal of the Korea Computer Graphics Society
• /
• v.23 no.3
• /
• pp.65-75
• /
• 2017

As a 360-degree image carries information of all directions, it often has too much information. Moreover, in order to investigate a 360-degree image on a 2D display, a user has to either click and drag the image with a mouse, or project it to a 2D panorama image, which inevitably introduces severe distortions. In consequence, investigating a 360-degree image and finding an object of interest in such a 360-degree image could be a tedious task. To resolve this issue, this paper proposes a method to find a region of interest and produces a 2D naturally looking image from a given 360-degree image that best matches a description given by a user in a natural language sentence. Our method also considers photo composition so that the resulting image is aesthetically pleasing. Our method first converts a 360-degree image to a 2D cubemap. As objects in a 360-degree image may appear distorted or split into multiple pieces in a typical cubemap, leading to failure of detection of such objects, we introduce a modified cubemap. Then our method applies a Long Short Term Memory (LSTM) network based object detection method to find a region of interest with a given natural language sentence. Finally, our method produces an image that contains the detected region, and also has aesthetically pleasing composition.

• Journal of the Korea Computer Graphics Society
• /
• v.21 no.2
• /
• pp.11-20
• /
• 2015

We present a new interactive technique for directing the motion sequences of an animated character by dragging its specific body part to a desired location in the three-dimensional virtual environment via a hand motion tracking device. The motion sequences of our character is synthesized by reordering subsequences of captured motion data based on a well-known graph representation. For each new input location, our system samples the space of possible future states by unrolling the graph into a spatial search tree, and retrieves one of the states at which the dragged body part of the character gets closer to the input location. We minimize the difference between each pair of successively retrieved states, so that the user is able to anticipate which states will be found by varying the input location, and resultantly, to quickly reach the desired states. The usefulness of our method is demonstrated through experiments with breakdance, boxing, and basketball motion data.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.6 s.44
• /
• pp.269-277
• /
• 2006

We propose a new expert system for recovering the broken fragments of relics into an original form using computer graphics and image processing. This paper presents a system with an application to tombstones objects of flat plane with letters carved in for assembling the fragments by placing their respective fragments in the right position. The matching process contains three sub-processes: aligning the front and letters of an object, identifying the matching directions, and determining the detailed matching positions. We apply least squares fitting, vector inner product, and geometric and RGB errors to the matching process. It turned out that 2-D translations via fragments-alignment enable us to save the computational load significantly. Based on experimental results from the damaged cultural fragments, the performance of the proposed method is illustrated.

• Journal of the Korea Computer Graphics Society
• /
• v.26 no.2
• /
• pp.1-9
• /
• 2020

In this paper, we propose haptic interfaces based on force feedback to provide a physical painting experience to virtual reality users. Through this system, the user can create surface-based painting holding a haptic stylus, while utilizing both visual feedback from the worn HMD and haptic feedback during painting. In particular, the haptic interfaces simulate the physical interaction between painting brush and painting, which can help to improve the spatial perception of users and compensate for visual feedback. This can reduce laborious drawing works to repeatedly paint strokes and therefore yield a better painting performance. As a result, users can experience more effective and realistic VR painting with this system.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.1
• /
• pp.57-62
• /
• 2006

This treatise proposes a develop system that manufactures 3D clothes model who is used in 3D clothes getting dressed simulation. Need 3D human body model and 3D clothes model to do getting dressed simulation. Create priority work Matrix javelin to design clothes model and design 2D piece from designer Connect designed piece plain using backstitch line and create numerical value data because using designed piece and backstitch data finally and make 3D clothes model. Consist of piece design module, to read clothes data and save module, getting dressed simulation module in system that propose in this treatise.