• Smart Media Journal
• /
• v.8 no.4
• /
• pp.64-71
• /
• 2019

The performance of an OMR (Optical Music Recognition) system is usually determined by the characterizing features of the input music score images. Low resolution is one of the main factors leading to degraded image quality. In this paper, we handle the low-resolution problem using the super-resolution technique. We propose the use of a deep neural network with instance normalization to improve the quality of music score images. We apply instance normalization which has proven to be beneficial in single image enhancement. It works better than batch normalization, which shows the effectiveness of shifting the mean and variance of deep features at the instance level. The proposed method provides an end-to-end mapping technique between the high and low-resolution images respectively. New images are then created, in which the resolution is four times higher than the resolution of the original images. Our model has been evaluated with the dataset "DeepScores" and shows that it outperforms other existing methods.

• Journal of Broadcast Engineering
• /
• v.19 no.2
• /
• pp.148-157
• /
• 2014

Various techniques have been developed for efficient compression of stereoscopic 3D videos. Mixed-resolution based approach is one representative bit-rate saving method based on the characteristics of human visual system that the mixed-resolution stereoscopic videos are perceived close to the higher resolution. However, when the difference between the left and right image resolutions is bigger than a certain threshold level, it causes the perceived quality degradation of the 3D images. Subsequently, several researches tried to find the correlation between the difference in resolution and the level of the perceived quality degradation, but they conducted the experiments just considering the difference in resolution without considering the viewing distances, so thereby different results were retrieved from test to test. In this work, we calculated the optimal viewing distance based on the human visual system, and conducted the subjective tests with the calculated viewing distance. With the results, we demonstrate that the fixed and mobile hybrid 3DTV, which is based on mixed-resolution stereoscopic images, can provide the high quality 3D services.

• Proceedings of the Optical Society of Korea Conference
• /
• 2004.02a
• /
• pp.248-251
• /
• 2004

근접장 초해상 기술 중 대표적인 Super-RENS (Super Resolution Near Field Structure) 기술을 이용하여 25GB 용량의 BD(Blu-ray Disc) 다음 세대로 예상되는 100 ${\sim}$ 200 GB 용량의 Media 및 기록 재생 기기를 개발하고자 한다. 현재까지 이 기술의 어려움이었던 Stability 특성은 10만회 재생 내구성을 확인하엿고 CNR (신호 대 잡음 비) 특성은 50 nm 미세 마크에 대하여 40 dB의 특성을 확보 하였다. 이 특성은 대략 75 GB에 대응하는 것으로 BD System을 그대로 사용하면서 달성한 세계 최고의 기록 밀도로 평가된다. 향후 과제는 Jitter. Error Rate등 실제 신호 특성의 확보로서, 이의 확보를 통하여 기술의 상품화 가능성이 확인될 것이다.

• Journal of Information Display
• /
• v.8 no.4
• /
• pp.27-31
• /
• 2007

In this paper, we present an efficient tiled stereo display system for tangible meeting. For tangible meeting, it is important to provide immersive display with high resolution image to cover up the field of view and provide to the local user the same environment as that of remote site. To achieve these, a high resolution image needs to be transmitted for reconstruction of remote world, and it should be displayed using a tiled display. However, it is hard to transmit high resolution image in real time due to the limit of network bandwidth, and so we receive multiple images and reconstruct a remote world with received images in advance. Then, we update only a specific area where remote user exists by receiving low resolution image in realtime. We synthesize the transmitted image to the existing environmental map of remote world and display it as a stereo image. For this, we developed a new system which supports GPU based real time warping and blending, automatic feature extraction using machine vision technique.

• Broadcasting and Media Magazine
• /
• v.25 no.2
• /
• pp.7-16
• /
• 2020

최근 Ultra-High Definition(UHD) 등의 고해상도 디스플레이가 시장에 등장하고 이에 따라 소비자의 요구가 커지면서 기존의 Full-High Definition(FHD)과 같은 저해상도(Low Resolution, LR) 영상을 고해상도(High Resolution, HR) 영상으로 변환할 수 있는 초해상화(Super-Resolution) 알고리즘에 대한 관심이 커지고 있다. 그러나 기존의 전통적인 초해상화 기법들은 고해상도 영상을 복원하는 과정에서 디테일한 부분의 화질 저화 및 열화가 존재하는 것을 확인할 수 있다. 본 논문에서는 이러한 문제를 해결하기 위해 최근 널리 연구되고 있는 딥러닝 기반의 초해상화 알고리즘 연구의 최신 기술 현황 및 동향을 소개하고자 한다. 딥러닝 기반의 초해상화 알고리즘은 기존의 전통적인 기법들에 비해 높은 성능을 보여주고 있으며 현재까지도 다양한 네트워크가 제안되며 활발히 연구되고 있다. 본 논문에서는 딥러닝 기반의 초해상화 알고리즘 중 대표적인 네트워크들을 분석하고 다양한 데이터 셋에 대한 해당 네트워크의 실험 결과를 통해 딥러닝 기반의 초해상화 알고리즘의 우수성을 확인하고자 한다.

• Journal of International Society for Simulation Surgery
• /
• v.1 no.2
• /
• pp.57-61
• /
• 2014

We present a robust 3D facial reconstruction method using a single image generated by face-specific super resolution technique. Based on the several consecutive frames with low resolution, we generate a single high resolution image and a three dimensional facial model based on it. To do this, we apply PME method to compute patch similarities for SR after two-phase warping according to facial attributes. Based on the SRI, we extract facial features automatically and reconstruct 3D facial model with basis which selected adaptively according to facial statistical data less than a few seconds. Thereby, we can provide the facial image of various points of view which cannot be given by a single point of view of a camera.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.17-25
• /
• 2020

In the image object classification problem, low-resolution images may have a negative impact on the classification result, especially when the classification method, such as a convolutional neural network (CNN) model, is trained on a high-resolution (HR) image dataset. In this paper, we analyze the behavior of applying a classical super-resolution (SR) method such as bicubic interpolation, and a deep CNN model such as SRCNN to enhance low-resolution (LR) weeds images used for classification. Using an HR dataset, we first train a CNN model for weeds image classification with a default input size of 128 × 128. Then, given an LR weeds image, we rescale to default input size by applying the bicubic interpolation or the SRCNN model. We analyze these two approaches on the Chonnam National University (CNU) weeds dataset and find that SRCNN is suitable for the image size is smaller than 80 × 80, while bicubic interpolation is convenient for a larger image.

• Journal of Broadcast Engineering
• /
• v.9 no.4 s.25
• /
• pp.391-401
• /
• 2004

In this paper, we introduce a novel coding method to efficiently enable spatial random access for high resolution video. In terms of resolution and display size, standard display devices (such as cathode-ray tubes. monitors. PDAs, and LCDs) do not sufficiently support high resolution video such as digital cinema and panoramic video. Currently, users have no choice but to view video at lower resolution as a result of down-sampling, or only a partial region of the video due to display size limitations. Our proposed method. which we call the B-map, represents the set of starting locations of the coded segments in a picture frame. This information, or B-map, is first sent to the decoder prior to the coded data stream of the frame and is then used for fast local decoding. To test our method, we compare our B-map with JPEG tiling and the JPEG Resynchronization marker. Experimental results show that the proposed coding method requires less overhead than existing methods during the same decoding time. The results show promise for future panoramic or digital cinema applications.

• ETRI Journal
• /
• v.44 no.1
• /
• pp.85-93
• /
• 2022

We propose an efficient method for synthesizing mesh-based realistic computer-generated hologram (CGH). In a previous nonanalytic mesh-based CGH synthesis, the angular spectrum of the two-dimensional (2D) plane is calculated using the fast Fourier transform (FFT) with the same size as the resolution of the final hologram. Because FFT increases the computation time as the size of the input matrix increases, the previous method has a problem: The higher the resolution of the hologram, the greater the computational load, thereby delaying synthesis time. In this study, when calculating the angular spectrum of the 2D plane in mesh-based CGH synthesis, we propose a method to calculate the angular spectrum by defining the 2D plane with an arbitrary size smaller than the resolution of the final hologram. The resolution adjustment method reduces the computation time and can be applied to occlusion culling and texturing for the realistic effect of mesh-based CGH. We describe the principle, error analysis, application of realistic effect, and experimental results of the proposed method.

• Journal of Broadcast Engineering
• /
• v.24 no.2
• /
• pp.299-305
• /
• 2019

In ISO MPEG, MPEG-H has been standardized to smoothly deliver huge volume of ultra-high resolution videos over heterogeneous network environments for the consumers. In the MPEG-H standard, MMT(MPEG Media Transport) protocol is the next generation MPEG media transport technology to provide hybrid media delivery over heterogeneous environments. It is important to sustain accurate media synchronization to provide ultra-high resolution video services over heterogeneous network environments. In this paper, we propose and verify an in-band media synchronization scheme to circumvent the problem of additional session setup and excessive traffic problems of the conventional out-of-band media synchronization supported by the MMT technology.