• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06c
• /
• pp.388-390
• /
• 2012

도시 지역의 영상에서는 가로수, 행인, 자동차 등 3차원 대상이 아닌 전경 물체가 빈번하게 나타남에 따라 건물 모델의 텍스처 생성에 어려움이 있다. 따라서 본 논문은 건물 영상에서 전경 물체에 의한 가림이 제거된 정명평행 텍스처를 생성하는 방법을 제안한다. 제안하는 방법은 사용자 조작을 이용하여 반자동으로 전경 물체에 의해 가려진 영역을 찾고, 이를 인페인팅 기법을 통해 채워넣음으로서 facade에 대한 텍스처를 생성한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.04a
• /
• pp.848-850
• /
• 2015

본 논문에서는 픽셀 그레디언트의 방향성 정보를 이용하여 복원하는 예제기반 인페인팅 방법을 제안한다. 제안하는 방법에서는 영상 내에서 픽셀 그레디언트의 방향성을 검출하여 강도가 높은 그레디언트를 가지는 픽셀들을 먼저 복원하는 특징을 가진다. 지역적인 그래디언트의 특성을 이용하여 복원 영역의 연결성을 높이고 에러의 파급 효과를 줄임으로써, 기존의 예제기반 인페인팅 방법보다 개선된 결과를 보여준다.

• Korean Journal of Remote Sensing
• /
• v.23 no.5
• /
• pp.439-445
• /
• 2007

Stereoscopic viewing is an efficient technique for not only computer vision but also remote sensing applications. Generally, stereo pair obtained at the same time is necessary for 3D viewing, but it is possible to synthesize a stereomate suitable for stereo view with a single image and disparity-map. There have been researches concerning the generation of the synthetic stereomate from remote sensing imagery. However it is hard to find researches concerning the restoration of occlusion in stereomate. In this paper, we generated synthetic stereomates from remote sensing images, focused on the occlusion restoration. In order to figure out proper restoration methods depending on the spatial resolution of remote sensing imagery, we tested several methods including general interpolation and inpainting technique, then evaluated the results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.524-543
• /
• 2022

Correct facade image parsing is essential to the semantic understanding of outdoor scenes. Unfortunately, there are often various occlusions in front of buildings, which fails many existing methods. In this paper, we propose an end-to-end deep network for facade parsing with occlusions. The network learns to decompose an input image into visible and invisible parts by occlusion reasoning. Then, a context aggregation module is proposed to collect nonlocal cues for semantic segmentation of the visible part. In addition, considering the regularity of man-made buildings, a repetitive pattern completion branch is designed to infer the contents in the invisible regions by referring to the visible part. Finally, the parsing map of the input facade image is generated by fusing the results of the visible and invisible results. Experiments on both synthetic and real datasets demonstrate that the proposed method outperforms state-of-the-art methods in parsing facades with occlusions. Moreover, we applied our method in applications of image inpainting and 3D semantic modeling.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.65-68
• /
• 2021

본 연구에서는 Dilated U-Net 기반의 이미지 복원기법을 통해 콘크리트 균열 추출 성능 개선 방안을 제안한다. 콘크리트 균열은 구조물의 미관상의 문제뿐 아니라 추후 큰 안전사고의 원인이 될 수 있어 초기대응이 중요하다. 현재는 점검자가 직접 육안으로 검사하는 외관 검사법이 주로 사용되고 있지만, 이는 정확성 및 비용, 시간, 그리고 안전성 면에서 한계를 갖고 있다. 이에 콘크리트 구조물 표면에 대해 획득한 영상 처리 기법을 사용한 검사 방식 도입의 관심이 늘어나고 있다. 또한, 딥러닝 기술의 발달로 딥러닝을 적용한 영상처리의 연구 역시 활발하게 진행되고 있다. 본 연구는 콘크리트 균열 추개선출 성능 개선을 위해 Dilated U-Net 기반의 이미지 복원기법을 적용하는 방안을 제안하였고 성능 검증 결과, 기존 U-Net 기반의 정확도가 98.78%, 조화평균 82.67%였던 것에 비해 정확도 99.199%, 조화평균 88.722%로 성능이 되었음을 확인하였다.

• Journal of the Korean Mathematical Society
• /
• v.61 no.3
• /
• pp.427-444
• /
• 2024

The matrix completion problem is to predict missing entries of a data matrix using the low-rank approximation of the observed entries. Typical approaches to matrix completion problem often rely on thresholding the singular values of the data matrix. However, these approaches have some limitations. In particular, a discontinuity is present near the thresholding value, and the thresholding value must be manually selected. To overcome these difficulties, we propose a shrinkage and thresholding function that smoothly thresholds the singular values to obtain more accurate and robust estimation of the data matrix. Furthermore, the proposed function is differentiable so that the thresholding values can be adaptively calculated during the iterations using Stein unbiased risk estimate. The experimental results demonstrate that the proposed algorithm yields a more accurate estimation with a faster execution than other matrix completion algorithms in image inpainting problems.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.6
• /
• pp.48-60
• /
• 2011

Nowadays, the 3D community is actively researching on 3D imaging and free-viewpoint video (FVV). The free-viewpoint rendering in multi-view video, virtually move through the scenes in order to create different viewpoints, has become a popular topic in 3D research that can lead to various applications. However, there are restrictions of cost-effectiveness and occupying large bandwidth in video transmission. An alternative to solve this problem is to generate virtual views using a single texture image and a corresponding depth image. A critical issue on generating virtual views is that the regions occluded by the foreground (FG) objects in the original views may become visible in the synthesized views. Filling this disocclusions (holes) in a visually plausible manner determines the quality of synthesis results. In this paper, a new approach for handling disocclusions using depth based inpainting algorithm in synthesized views is presented. Patch based non-parametric texture synthesis which shows excellent performance has two critical elements: determining where to fill first and determining what patch to be copied. In this work, a noise-robust filling priority using the structure tensor of Hessian matrix is proposed. Moreover, a patch matching algorithm excluding foreground region using depth map and considering epipolar line is proposed. Superiority of the proposed method over the existing methods is proved by comparing the experimental results.

• Journal of Conservation Science
• /
• v.27 no.2
• /
• pp.191-199
• /
• 2011

A study was done to compare the properties of artificially deteriorated silk with ultraviolets for reinforcing of loss area of paintings on silk. Deteriorated surface of raw silk irradiated by long-wavelength ultraviolet(UV-A) than short-wavelength ultraviolet(UV-C) was similar to naturally aged raw silk. UV-A irradiation raw silk was slowly decreased in tensile tenacity and elongation and lowered in yellowness index than that of UV-C. Water content of UV-A irradiation raw silk than that of UV-C was higher. UV-A irradiation raw silk had no problem in dyeing and inpainting for conservation because of low yellowness index. UV-C irradiation raw silk was brittle, but UV-A irradiation raw silk was seemed to tough and similar to naturally aged raw silk. Korean painting conservator estimated that UV-A irradiation raw silk was more proper for reinforcing of loss area of paintings on silk than that of UV-C.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.3
• /
• pp.234-240
• /
• 2021

In this paper, the mask-based speech enhancement is improved for effective speech recognition in noise environments. In the mask-based speech enhancement, enhanced spectrum is obtained by multiplying the noisy speech spectrum by the mask. The VoiceFilter (VF) model is used as the mask estimation, and the Spectrogram Inpainting (SI) technique is used to remove residual noise of enhanced spectrum. In this paper, we propose a combined loss to further improve speech enhancement. In order to effectively remove the residual noise in the speech, the positive part of the Triplet loss is used with the component loss. For the experiment TIMIT database is re-constructed using NOISEX92 noise and background music samples with various Signal to Noise Ratio (SNR) conditions. Source to Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short-Time Objective Intelligibility (STOI) are used as the metrics of performance evaluation. When the VF was trained with the mean squared error and the SI model was trained with the combined loss, SDR, PESQ, and STOI were improved by 0.5, 0.06, and 0.002 respectively compared to the system trained only with the mean squared error.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.9
• /
• pp.4405-4418
• /
• 2016

Automatic image completion techniques have difficulty processing images in which the target region has multiple planes or is non-facade. Here, we propose a new image completion method that uses belief propagation based on planar priorities. We first calculate planar information, which includes planar projection parameters, plane segments, and repetitive regularity extractions within the plane. Next, we convert this planar information into planar guide knowledge using the prior probabilities of patch transforms and offsets. Using the energy of the discrete Markov Random Field (MRF), we then define an objective function for image completion that uses the planar guide knowledge. Finally, in order to effectively optimize the MRF, we propose a new optimization scheme, termed Planar Priority-belief propagation that includes message-scheduling-based planar priority and dynamic label cropping. The results of experiment show that our approach exhibits advanced performance compared with existing approaches.