Joohyeok Kim

Joint-adaptive bilateral depth map upsampling

Abstract In this paper, we present a new depth upsampler in which an upsampled depth map is computed at each pixel as the average of neighboring pixels, weighted by color and depth intensity filters. The proposed method features two parameters, an adaptive smoothing parameter and a control parameter. The adaptive smoothing parameter is determined based on the ratio between a depth map and its corresponding color image. The adaptive smoothing parameter is used to control the dynamic range of the color-range filter. The control parameter assigns a larger weighting factor to pixels in the object to which a missing pixel belongs. In a comparison with five existing upsamplers, the proposed method outperforms all five in terms of both objective and subjective quality.

Fast Intra Mode Decision Algorithm Using the Sum of Absolute Transformed Differences

This paper presents a fast intra mode decision algorithm using the sum of absolute transformed differences (SATD) in order to reduce the encoding time. Our algorithm consists of two main techniques: Early block-mode decision and candidate directional-mode selection. First, we use the early block-mode decision method, I4MB skip, to avoid unnecessary computation for I4MB. If all the 16 4x4sub-blocks in intra 16x16block (I16MB) have the minimum SATD in the same directional-mode, then we skip the I4MB prediction and choose the I16MB as the best block-mode for the current block. Besides, before performing the rate-distortion cost for all nine directional-modes of I4MB, which brings out the huge complexity, we reduce the number of the candidate directional-modes by comparing the SATD of each mode with the SATD of primary prediction mode (PPM). The PPM is obtained from the combination of the best mode of I16MB and the modes of neighboring blocks. Experimental results indicate that the proposed algorithm saved encoding time about 70% with negligible degradation.

Taylor series and adaptive fusion strategy for Bayer demosaicking

Demosaicking is an important step in generating a full-color image from a color filter array image captured by a single-sensor digital camera. In this paper, we propose a new demosaicking algorithm which uses Taylor series and an adaptive fusion strategy. For generating directional predictors, we present a new interpolator based on Taylor series. The directional predictors are used together with spatial and spectral correlations. Using these predictors and the adaptive fusion strategy improves the green channel interpolation, in turn helping to reconstruct the red and blue channel interpolations. Experiments comparing the proposed algorithm to some other demosaicking algorithms show that the proposed algorithm provides superior performance in terms of both objective and subjective image quality.

A New Adaptive Linear Interpolation Algorithm Using Pattern Weight Based on Inverse Gradient

Most of conventional interpolation methods do not consider the sufficient pattern of neighbor pixels, which cause quality degradation. Kim et al. proposed New Adaptive Linear (NAL) algorithm to consider patterns near the interpolated value [8]. However, they have a critical defect which does not reflect whether each neighbor pixels influence the interpolated pixel. To remove this defect, we propose a new image interpolation method using adaptive weight based on inverse gradient. Experimental results show that the proposed algorithm exhibits a better performance than conventional algorithms in both objective and subjective criteria with a variety of images. In addition, the proposed method just needs a little computational burden compared with other algorithms.

Low-complexity block-based motion estimation algorithm using adaptive search range adjustment

We propose an adaptive search range adjustment algorithm that is particularly suitable for low-bit-depth motion estimation approaches. Typical low-bit-depth motion estimation approaches, such as one-bit transform (1BT), two-bit transform (2BT), and truncated gray-coded bit-plane matching (TGCBM), perform a full search with low-bit expression. We propose combining a search-range adjustment, which can be predicted by using motion vector information of the reference frame and the existing bit-wise matching methods, such as 1BT, 2BT, and TGCBM. Combining this proposed algorithm and low-bit expression enhances both motion estimation accuracy and time consumption. Experimental results show that low-bit-depth motion estimation approaches using this proposed algorithm have an outstanding performance compared with those without the proposed algorithm both in motion estimation accuracy and time consumption.

Video Sequence Deinterlacing Using Intensity Gradient Filter and Median Filter with Texture Detection

In this paper, we propose a new de-interlacing algorithm for video data using intensity gradient filter and median filter with texture detection in the image block. We first introduce the texture detection. According to texture detection, the current region is determined into smooth region or texture region. In case that the smooth region interpolated by median filter. In addition, in case of the texture region, we calculate missing pixel value using intensity gradient filter. Therefore, we analyze the local region feature using the texture detection and classify each missing pixel into two categories. And then, based on the classification result, a different deinterlacing algorithm is activated in order to obtain the best performance. Experimental results show that the proposed algorithm performs well with a variety of moving sequences compared with traditional intra-field method in the literature.

New edge-guided interpolation algorithm based on Weighted Edge Pattern Analysis

Most of previous interpolation methods cannot consider the sufficient edge information, and that result in unwanted artifacts. To remove or reduce the problem effectively, we propose a new image interpolation scheme based on Weighted Edge Pattern Analysis (WEPA). Proposed algorithm finds the accurate edge direction at current pixel by using the Spatial Direction Vector (SDV) which is result of WEPA of the neighboring pixels. After direction refinement, we interpolate the current pixel along obtained final SDV direction. Experimental results show that the proposed method exhibits a better performance than previous methods in both objective and subjective results with a variety of still images.

Autocorrelation-based interlaced to progressive format conversion

Abstract To generate a high resolution image from a low resolution one, interpolation plays a crucial role. However, conventional interpolation methods including edge-based interpolation methods have some drawbacks such as the limited number of edge directions, imprecise edge detection, and inefficient interpolation. To overcome these shortcomings, we propose a new edge-directed interpolation method, which has three aims: various edge directions, reliable edge detection, and outstanding interpolation. Since the number of candidate edge directions in the proposed method is flexible, we can use several edges included in the high resolution image. To accurately determine the edge direction, we use autocorrelation of neighboring pixels for candidate directions based on the duality between a high resolution image and its corresponding low resolution image. For the interpolation step, we utilize a Blackman–Harris windowed-sinc weighted average filter where we use correlation values obtained in the edge detection step as weights. Experimental results show that the proposed method outperforms conventional methods in terms of both the subjective and the objective results.