• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.7
• /
• pp.4545-4553
• /
• 2014

Mobile GPU has led to the rapid development of smart phone graphic technology. Most recent smart phones are equipped with high-performance multi-core GPU. How a multi-core mobile GPU can be utilized efficiently will be a critical issue for improving the smart phone performance. On the other hand, most current research has focused on a single-core mobile GPU; studies of multi-core mobile GPU are rare. In this paper, the job scheduling patterns and the efficiency of multi-core mobile GPU are analyzed. In the profiling result, despite the higher number of GPU cores, the total processing time required for certain graphics applications were increased. In addition, when GPU is processing for 3D games, a substantial amount of overhead is caused by communication between not only the CPU and GPU, but also within the GPUs. These results confirmed that more active research for multi-core mobile GPU should be performed to optimize the present mobile GPUs.

• International Journal of Computer Science & Network Security
• /
• v.22 no.9
• /
• pp.175-182
• /
• 2022

In our daily life, we come across different types of information, for example in the format of multimedia and text. We all need different types of information for our common routines as watching/reading the news, listening to the radio, and watching different types of videos. However, sometimes we could run into problems when a certain type of information is required. For example, someone is listening to the radio and wants to listen to jazz, and unfortunately, all the radio channels play pop music mixed with advertisements. The listener gets stuck with pop music and gives up searching for jazz. So, the above example can be solved with an automatic audio classification system. Deep Learning (DL) models could make human life easy by using audio classifications, but it is expensive and difficult to deploy such models at edge devices like nano BLE sense raspberry pi, because these models require huge computational power like graphics processing unit (G.P.U), to solve the problem, we proposed DL model. In our proposed work, we had gone for a low complexity model for Audio Event Detection (AED), we extracted Mel-spectrograms of dimension 128×431×1 from audio signals and applied normalization. A total of 3 data augmentation methods were applied as follows: frequency masking, time masking, and mixup. In addition, we designed Convolutional Neural Network (CNN) with spatial dropout, batch normalization, and separable 2D inspired by VGGnet [1]. In addition, we reduced the model size by using model quantization of float16 to the trained model. Experiments were conducted on the updated dataset provided by the Detection and Classification of Acoustic Events and Scenes (DCASE) 2020 challenge. We confirm that our model achieved a val_loss of 0.33 and an accuracy of 90.34% within the 132.50KB model size.

• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.849-858
• /
• 2019

Recent developments in computer graphics technology and image processing technology have increased interest in point cloud technology for inputting real space and object information as three-dimensional data. In particular, point cloud technology can accurately provide spatial information, and has attracted a great deal of interest in the field of autonomous vehicles and AR (Augmented Reality)/VR (Virtual Reality). However, in order to provide users with 3D point cloud contents that require more data than conventional 2D images, various technology developments are required. In order to solve these problems, an international standardization organization, MPEG(Moving Picture Experts Group), is in the process of discussing efficient compression and transmission schemes. In this paper, we provide a region division transfer method of 3D point cloud content through extension of existing MPEG-DASH (Dynamic Adaptive Streaming over HTTP)-SRD (Spatial Relationship Description) technology, quality parameters are further defined in the signaling message so that the quality parameters can be selectively determined according to the user's request. We also design a verification platform for ROUTE (Real Time Object Delivery Over Unidirectional Transport)/DASH based heterogeneous network environment and use the results to validate the proposed technology.

• Journal of Korea Game Society
• /
• v.9 no.4
• /
• pp.89-96
• /
• 2009

In general, the size of volume data is large since it has logical 3D structure so it takes long time to manipulate. Much work has been done to improve processing speed of volume data. In this paper, we propose a interlaced scanning volume rendering that reduce computation time by using temporal coherence with minimum loss of image quality. It renders a current frame by reusing information of previous frame. Conventional volume raycasting renders each frame by casting rays on every pixels. On the other hand, our methods divided an image into n-pixel blocks, then it casts a ray on a pixel of a block per each frames. Consequently, it generates an image by accumulating pixel values of previous n frames. The quality of rendered image of our method is better than that of simple screen space subsampling method since it uses afterimage effect of human cognitive system, and it is n-times faster that the previous one.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.7
• /
• pp.340-353
• /
• 2008

A computational color palette extraction model is introduced to describe paint brush objectively and efficiently. In this model, a color palette is defined as a minimum set of colors in which a painting can be displayed within error allowance and extracted by the two step processing of color grouping and major color extraction. The color grouping controls the resolution of colors adaptively and produces a basic color set of given painting images. The final palette is obtained from the basic color set by applying weighted k-means clustering algorithm. The extracted palettes from several famous painters are displayed in a 3-D color space to show the distinctive palette styles using RGB and CIE LAB color models individually. And the two experiments of painter classification and color transform of photographic image has been done to check the performance of the proposed method. The results shows the possibility that the proposed palette model can be a computational color analysis metric to describe the paint brush, and can be a color transform tool for computer graphics.

• Journal of Korea Multimedia Society
• /
• v.10 no.11
• /
• pp.1460-1471
• /
• 2007

A 3-Dimensional engine in a mobile embedded device is divided into a C-based OpenGL/ES and a Java-based JSR184 which interprets and executes a byte code in a real-time. In these two standards, the JSR184 supporting Java objects uses more processor resources than an OpenGL/ES and thus has a constraint when it is used in an embedded device with a limited computing power. On the other hand, 3-Dimensional contents employed in existing personal computer are created by utilizing advantages of Java and secured numerous users in European market, due to the good quality in contents and extensive service in a commercial network, GSM. Because of the reason, a mobile embedded device used in a GSM network needs a JSR184 which can provide an existing Java-based 3-Dimensional contents without extra conversion processes, but the current version of Java-based 3-Dimensional engine has drawbacks in application to commercial products because it requires more computing power than the mobile embedded device. This paper proposes a binding technique with the advantages of Java objects to improve a processing speed of 3-Dimensional contents in limited resources of a mobile embedded device. The technique supports a JSR184 standard interface in the upper layer to utilize 3-Dimensional contents using Java, employs a different code-conversion language, KNI(Kilo Native Interface), in the middle layer to interface between OpenGL/ES and JSR184, and embodies an OpenGL/ES standard in the lower layer. The validity of the binding technique is demonstrated through a simulator and a FPGA embedding an ARM.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.351-363
• /
• 2023

Visual structural inspections are an inseparable part of post-earthquake damage assessments. With unmanned aerial vehicles (UAVs) establishing a new frontier in visual inspections, there are major computational challenges in processing the collected massive amounts of high-resolution visual data. We propose twin deep learning models that can provide accurate high-resolution structural components and damage segmentation masks efficiently. The traditional approach to cope with high memory computational demands is to either uniformly downsample the raw images at the price of losing fine local details or cropping smaller parts of the images leading to a loss of global contextual information. Therefore, our twin models comprising Trainable Resizing for high-resolution Segmentation Network (TRS-Net) and DmgFormer approaches the global and local semantics from different perspectives. TRS-Net is a compound, high-resolution segmentation architecture equipped with learnable downsampler and upsampler modules to minimize information loss for optimal performance and efficiency. DmgFormer utilizes a transformer backbone and a convolutional decoder head with skip connections on a grid of crops aiming for high precision learning without downsizing. An augmented inference technique is used to boost performance further and reduce the possible loss of context due to grid cropping. Comprehensive experiments have been performed on the 3D physics-based graphics models (PBGMs) synthetic environments in the QuakeCity dataset. The proposed framework is evaluated using several metrics on three segmentation tasks: component type, component damage state, and global damage (crack, rebar, spalling). The models were developed as part of the 2^nd International Competition for Structural Health Monitoring.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.3
• /
• pp.197-205
• /
• 2020

Korean language has the characteristics that the pronunciation of phoneme units such as vowels and consonants are fixed and the pronunciation associated with a notation does not change, so that foreign learners can approach rather easily Korean language. However, when one pronounces words, phrases, or sentences, the pronunciation changes in a manner of a wide variation and complexity at the boundaries of syllables, and the association of notation and pronunciation does not hold any more. Consequently, it is very difficult for foreign learners to study Korean standard pronunciations. Despite these difficulties, it is believed that systematic analysis of pronunciation errors for Korean words is possible according to the advantageous observations that the relationship between Korean notations and pronunciations can be described as a set of firm rules without exceptions unlike other languages including English. In this paper, we propose a visualization framework which shows the differences between standard pronunciations and erratic ones as quantitative measures on the computer screen. Previous researches only show color representation and 3D graphics of speech properties, or an animated view of changing shapes of lips and mouth cavity. Moreover, the features used in the analysis are only point data such as the average of a speech range. In this study, we propose a method which can directly use the time-series data instead of using summary or distorted data. This was realized by using the deep learning-based technique which combines Self-organizing map, variational autoencoder model, and Markov model, and we achieved a superior performance enhancement compared to the method using the point-based data.

• Journal of Korea Water Resources Association
• /
• v.47 no.12
• /
• pp.1145-1154
• /
• 2014

In recent years, as frequency and intensity of severe weather disasters such as flash flood have been increasing, providing accurate and prompt information to the public is very important and needs of user-friendly monitoring/warning system are growing. This paper introduces a method that re-produces radar observations as multimedia contents and applies reproduced data to mesh-up services. In addition, a accurate GIS matching technique to help to track the exact location going on serious atmospheric phenomena is presented. The proposed method create multimedia contents having structures such as two dimensional images, vector graphics or three dimensional volume data by re-producing various radar variables obtained from a weather radar. After then, the multimedia formatted weather radar data are matched with various detailed raster or vector GIS map platform. Results of simulation test with various scenarios indicate that the display system based on the proposed method can support for users to figure out easily and intuitively routes and degrees of risk of severe weather. We expect that this technique can also help for emergency manager to interpret radar observations properly and to forecast meteorological disasters more effectively.

• Cartoon and Animation Studies
• /
• s.45
• /
• pp.197-214
• /
• 2016

Augmented Reality, virtual reality in recently attracted attention throughout the world. and Mix them mixed reality etc., it has had a significant impact on the overall pop culture beyond the scope of science and technology. The world's leading IT company : Google, Apple, Samsung, Microsoft, Sony, LG is focusing on development of AR, VR technology for the public. The many large and small companies developed VR hardware, VR software, VR content. It does not look that makes a human a human operation in the cognitive experience of certain places or situations or invisible through Specific platforms or program is Encompass a common technique that a realization of the virtual space. In particular, out of the three-dimensional image reveals the limitations of the conventional two-dimensional structure - 180, 360 degree images provided by the subjective and objective symptoms such as vision and sense of time and got participants to select it. VR technology that can significantly induce the commitment and participation is Industry as well as to the general public which leads to the attention of colostrum. It was introduced more than 10 related VR works Year 2015 Sundance Film Festival New Frontier program. The appearance VR content : medical, architecture, shopping, movies, animations. Also, 360 individuals can be produced by the camera / video sharing VR is becoming an interactive tunnel between two possible users. Nevertheless, This confusion of values, moral degeneration and the realization of a virtual space that has been pointed out that the inherent. 4K or HUD, location tracking, motion sensors, processing power, and superior 3D graphics, touch, smell, 4D technology, 3D audio technology - It developed more than ever and possible approaches to reality. Thereafter, This is because the moral degeneration, identity, generational conflict, and escapism concerns. Animation is also seeking costs in this category Reality. Despite the similarities rather it has that image, and may be the reason that the animation is pushed back to the VR content creation. However, it is focused on the game and VR technology and the platform that is entertaining, but also seek new points within the animation staying in the flat Given that eventually consist of visual images is clear that VR sought. Finally, What is the reality created in the virtual space using VR technology could be applied to the animation? So it can be seen that the common interest is research on what methods and means applied.