• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.808-812
• /
• 2003

The technology and concept of multimedia expands with the rapid growth of Digital technology. Digital broadcasting has already begun and movie industry is also switching the way of visualization from analog to digital. Image compositing is the core of the computer graphics related multimedia technology. Currently, various computer graphic technologies are being developed. Nevertheless, image compositing are being done manually because of not only the expensiveness image compositing equipments, but also the lack of experts in this area. This paper present a new image compositing technique for 3-Dimensional graphics and digital sources. A 2-Dimensional subject which will be used as a background was expanded into a 3-Dimensional by using the tracking technique. Thereafter, Effective image compositing has been obtained by precisely cooperating digital source images and 3-Dimensional model using 3-Dimensional graphics editing tool.

• Journal of KIISE
• /
• v.42 no.6
• /
• pp.699-706
• /
• 2015

In recent years, as near-duplicate image has been increasing explosively by the spread of Internet and image-editing technology that allows easy access to image contents, related research has been done briskly. However, BoF (Bag-of-Feature), the most frequently used method for near-duplicate image detection, can cause problems that distinguish the same features from different features or the different features from same features in the quantization process of approximating a high-level local features to low-level. Therefore, a post-verification method for BoF is required to overcome the limitation of vector quantization. In this paper, we proposed and analyzed the performance of a post-verification method for BoF, which converts SIFT (Scale Invariant Feature Transform) descriptors into 128 bits binary codes and compares binary distance regarding of a short ranked list by BoF using the codes. Through an experiment using 1500 original images, it was shown that the near-duplicate detection accuracy was improved by approximately 4% over the previous BoF method.

• Journal of the Korea Computer Graphics Society
• /
• v.21 no.4
• /
• pp.1-10
• /
• 2015

Recently, calligraphy has become popular because people focused on emotion. The strokes, dots, swoops, cracks and shading are the calligraphy factors for expressing various emotions such as joy, anger, sorrow, and delight. However, the emotion which is expressed by cracks and shading can be destroyed in the digital work when the calligraphy is used for a variety sizes of prints. Professionals work with high-resolution images which are obtained through the scanner, however normal users should work with low-resolution images taken with the smart phone for calligraphy image editing. We propose a raster image scaling method focused on calligraphy that maintains the emotion with cracks and shading, when normal users use the low-resolution calligraphy images in the digital work. The method recolors aliasing boundary of enlarged rasterized image. When recolored by our method, our method decreases aliasing by using the image gradient method, vivify calligraphy images, and maintains the emotion in cracks and shading by using the alpha value.

• Journal of Digital Convergence
• /
• v.17 no.4
• /
• pp.255-260
• /
• 2019

This study is about an efficient synthesis of $360^{\circ}$ video and 3D graphics. First, the video image filmed by a binocular integral type $360^{\circ}$ camera was stitched, and location values of the camera and objects were extracted. And the data of extracted location values were moved to the 3D program to create 3D objects, and the methods for natural compositing was researched. As a result, as the method for natural compositing of $360^{\circ}$ video image and 3D graphics, rendering factors and rendering method were derived. First, as for rendering factors, there were 3D objects' location and quality of material, lighting and shadow. Second, as for rendering method, actual video based rendering method's necessity was found. Providing the method for natural compositing of $360^{\circ}$ video image and 3D graphics through this study process and results is expected to be helpful for research and production of $360^{\circ}$ video image and VR video contents.

• Journal of Information Processing Systems
• /
• v.19 no.1
• /
• pp.67-79
• /
• 2023

In the image field, data augmentation refers to increasing the amount of data through an editing method such as rotating or cropping a photo. In this study, a generative adversarial network (GAN) image was created using CycleGAN, and various colors of dogs were reflected through data augmentation. In particular, dog data from the Stanford Dogs Dataset and Oxford-IIIT Pet Dataset were used, and 10 breeds of dog, corresponding to 300 images each, were selected. Subsequently, a GAN image was generated using CycleGAN, and four learning groups were established: 2,000 original photos (group I); 2,000 original photos + 1,000 GAN images (group II); 3,000 original photos (group III); and 3,000 original photos + 1,000 GAN images (group IV). The amount of data in each learning group was augmented using existing data augmentation methods such as rotating, cropping, erasing, and distorting. The augmented photo data were used to train the MobileNet_v3_Large, ResNet-152, InceptionResNet_v2, and NASNet_Large frameworks to evaluate the classification accuracy and loss. The top-3 accuracy for each deep neural network model was as follows: MobileNet_v3_Large of 86.4% (group I), 85.4% (group II), 90.4% (group III), and 89.2% (group IV); ResNet-152 of 82.4% (group I), 83.7% (group II), 84.7% (group III), and 84.9% (group IV); InceptionResNet_v2 of 90.7% (group I), 88.4% (group II), 93.3% (group III), and 93.1% (group IV); and NASNet_Large of 85% (group I), 88.1% (group II), 91.8% (group III), and 92% (group IV). The InceptionResNet_v2 model exhibited the highest image classification accuracy, and the NASNet_Large model exhibited the highest increase in the accuracy owing to data augmentation.

• Cartoon and Animation Studies
• /
• s.48
• /
• pp.117-139
• /
• 2017

The possibility of interactivity of digital media environment is adopted in human expression system and integrates the dynamic aspect of digital technology with expressive structure, thereby transforming the paradigm of image acceptance as well as image expression range. Virtual reality images have an important meaning in that they are changing the one-way mechanism of production and acceptance of images that lead to producers-video-audiences beyond the problem of verisimilitude such as how vividly they simulate reality. First of all, the virtual reality image is not one-sided but interactive image composed by the user. Viewing a virtual reality image does not just see the camera shine, but it gets the same view as in the real world. Therefore, the image that was controlled through framing changes to be configured positively by the user. This implies a change in the paradigm of image acceptance as well as a change in the existing form of the image itself. In addition, the narrative structure of the image and the subjects that are formed in the process are also required to be discussed. In the virtual reality image, the user 's gaze is a fusion of the gaze inside the image and the gaze outside the image. This is because the position of the user as the subject of the gaze in the virtual reality image is continuously restricted by the device of the discourse such as the editing and the narration of the shot. The significance of the virtual reality image is not aesthetically perfect but it is reconstructed according to the user to reflect the existence of the user positively and engage the user in the image.

• The KIPS Transactions:PartB
• /
• v.11B no.4
• /
• pp.527-536
• /
• 2004

TheReviser is a digital document revision application on a projection display, which allows us to interact a digital document with the same gestures used for paper documents revision. To enable these interactions, TheReviser should detect foreground objects such as hands or pens on a projection display, and should spot and recognize gesture commands from continuous movements of a user. To detect foreground objects from a complex background in various lighting conditions, we perform geometry and color calibration between a captured image and a frame buffer image. TheReviser uses an HMM-based gesture recognition method Experimental results show that the proposed application recognizes user's gestures on average 93.22% in test gesture sequences.

• Proceedings of the Korea Contents Association Conference
• /
• 2005.05a
• /
• pp.165-171
• /
• 2005

we can define the procedure of Authoring that it makes area cord and the reproduction prevent menu programmed into MPEG II video stream , Audio which is AC-3 audio stream and subtitle under its own category. And it makes process an attribute, an order and an operation, gives the last disk image (DVD). There are various types of Authoring tools in the market so that authoring tools can enable, encourage, and assist users ('authors') in the selection of tools that produce simple title and video production and editing suites. In this paper, we will compare and analyze authoring process in which image and sound are captured into DVD creation with IEEE 1394port with regard to Window system using generally with Desktop PC and the Macintosh of the PC on the system of Windows and OSX.

• The Journal of the Korea Contents Association
• /
• v.5 no.5
• /
• pp.145-151
• /
• 2005

We can define the procedure of Authoring that it makes area cord and the reproduction prevent menu programmed into MPEG II video stream , Audio which is AC-3 audio stream and subtitle under its own category. And it makes process an attribute, an order and an operation, gives the last disk image, Which is DVD(digital versatile disc). There are various process of Authoring tools in the market so that authoring tools can enable, encourage, and assist users ('authors') in the selection of tools that produce simple title, video production and editing suites. In this paper, we will compare and analyze authoring process in which image and sound are authorized into DVD with IEEE 1394port between Window system using generally with Desktop PC and the Macintosh that is based on OSX.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.4300-4314
• /
• 2019

With the development of video-related applications, media content has increased dramatically through applications. There is a substantial amount of near-duplicate videos (NDVs) among Internet videos, thus NDVR is important for eliminating near-duplicates from web video searches. This paper proposes a novel NDVR system that supports large-scale retrieval and contributes to the efficient and accurate retrieval performance. For this, we extracted keyframes from each video at regular intervals and then extracted both commonly used features (LBP and HSV) and new image features from each keyframe. A recent study introduced a new image feature that can provide more robust information than existing features even if there are geometric changes to and complex editing of images. We convert a vector set that consists of the extracted features to binary code through a set of hash functions so that the similarity comparison can be more efficient as similar videos are more likely to map into the same buckets. Lastly, we calculate similarity to search for NDVs; we examine the effectiveness of the NDVR system and compare this against previous NDVR systems using the public video collections CC_WEB_VIDEO. The proposed NDVR system's performance is very promising compared to previous NDVR systems.