Bilateral Filtering

Depth maps aid the generation of virtual viewpoints in multiview video applications. As the depth maps can be represented as a gray scale image sequence, they can be compressed using existing video codecs. However, when the depth maps are compressed using existing codecs, compression artifacts cause undesirable distortions in the rendered views. If the effects of compression artifacts can be minimized at the decoder, it would be possible to use the existing video codecs, which achieve significant compression ratios, to effectively compress depth maps. In this paper we theoretically analyze the effects of compression artifacts on the virtual view generation process. Based on this analysis, a set of guidelines are formulated for designing a post-processing filter, which could effectively minimize the effects of compression artifacts. Assuming that depth maps are piecewise smooth images with sharp discontinuities, an adaptive bilateral filtering technique is proposed as an out-of-the-loop filter at the decoder to post-process the compressed depth maps. Histograms of the compressed depth maps are analyzed on block basis to identify the dominant depth value bins in each block. Using the identified global depth value bins, the proposed technique successfully minimizes the artifacts by adjusting the histograms of the compressed depth maps. The experimental results suggest that the proposed depth map filtering technique can significantly improve the average perceptual quality of rendered views
by up to 1.7dB over the state-of-the-art techniques. It is expected that the proposed post-processing technique will have important use cases in advanced 3-Dimensional Television systems.

Although three-dimensional (3D) displays enhance visual quality more than two-dimensional (2D) displays do, the depth information required for 3D displays is unavailable in the conventional 2D content. Therefore, converting 2D videos into 3D ones has become an important issue in emerging 3D applications. This work presents a novel algorithm that automatically converts 2D videos into 3D ones. The proposed algorithm utilizes the edge information to segment the image into object groups. A depth map is then assigned based on a hypothesized depth gradient model. Next, the depth map is block-based assigned by cooperating with a cross bilateral filter to generate visually comfortable depth maps efficiently and also diminish the block artifacts. A multiview video can be readily generated by using a depth image-based rendering method.

This paper presents a new algorithm for denoising dynamic contrast-enhanced (DCE) MR images. It is a novel variation on the nonlocal means (NLM) algorithm. The algorithm, called dynamic nonlocal means (DNLM), exploits the redundancy of information in the temporal sequence of images. Empirical evaluations of the performance of the DNLM algorithm relative to seven other denoising methods-simple Gaussian filtering, the original NLM algorithm, a trivial extension of NLM to include the temporal dimension, bilateral filtering, anisotropic diffusion filtering, wavelet adaptive multiscale products threshold, and traditional wavelet thresholding-are presented. The evaluations include quantitative evaluations using simulated data and real data (20 DCE-MRI data sets from routine clinical breast MRI examinations) as well as qualitative evaluations using the same real data (24 observers: 14 image/signal-processing specialists, 10 clinical breast MRI radiographers). The results of the quantitative evaluation using the simulated data show that the DNLM algorithm consistently yields the smallest MSE between the denoised image and its corresponding original noiseless version. The results of the quantitative evaluation using the real data provide evidence, at the ?? = 0.05 level of significance, that the DNLM algorithm yields the smallest MSE between the denoised image and its corresponding original noiseless version. The results of the qualitative evaluation provide evidence, at the ?? = 0.05 level of significance, that the DNLM algorithm performs visually better than all of the other algorithms. Collectively the qualitative and quantitative results suggest that the DNLM algorithm more effectively attenuates noise in DCE MR images than any of the other algorithms.

This paper presents a new algorithm for denoising dynamic contrast-enhanced (DCE) MR images. It is a novel variation on the nonlocal means (NLM) algorithm. The algorithm, called dynamic nonlocal means (DNLM), exploits the redundancy of information in the temporal sequence of images. Empirical evaluations of the performance of the DNLM algorithm relative to seven other denoising methods-simple Gaussian filtering, the original NLM algorithm, a trivial extension of NLM to include the temporal dimension, bilateral filtering, anisotropic diffusion filtering, wavelet adaptive multiscale products threshold, and traditional wavelet thresholding-are presented. The evaluations include quantitative evaluations using simulated data and real data (20 DCE-MRI data sets from routine clinical breast MRI examinations) as well as qualitative evaluations using the same real data (24 observers: 14 image/signal-processing specialists, 10 clinical breast MRI radiographers). The results of the quantitativ...

Boostrap is a statistical method widely used in
applications such as model averaging and noise estimation. In
this paper, we propose a bilateral smoothing algorithm based on
bootstrap noise estimates for denoising 3D point sets. Following
a classic denoising technique, for a given neighbourhood of the
point set, we first fit a polynomial surface and then update
the position of each point of the neighbourhood by moving
it towards its projection on the fitted surface. However, in the
proposed algorithm, the amount by which we move each vertex
towards its projection depends on the bootstrap error estimate
of the polynomial fitting. As a result, low quality polynomial
fittings with high bootstrap error estimates tend have smaller
effect on the denoising process and do not degrade the final
result. We also propose a multi-pass, density adaptive variant
of the proposed bilateral smoothing and experimentally show
that it can further improve the results.

We provide and evaluate a fusion algorithm of remotely sensed images, i.e. the fusion of a panchromatic (PAN) image with a multi-spectral (MS) image using bilateral filtering, applied to images of three different sensors: SPOT 5, Landsat ETM+ and Quickbird. To assess the fusion process, we use six quality indexes, that confirm, along with visual analysis, good overall results for the three sensors.

Filtering has been a major approach used in the homogenization-based methods for structural topology optimization to suppress the checkerboard pattern and relieve the numerical instabilities. In this paper a bilateral filtering technique originally developed in image processing is presented as an efficient approach to regularizing the topology optimization problem. A non-linear bilateral filtering process leads to a suitable problem regularization to eliminate the checkerboard instability, pronounced edge preserving smoothing characteristics to favour the 0–1 convergence of the mass distribution, and computational efficiency due to its single pass and non-iterative nature. Thus, we show that the application of the bilateral filtering brings more desirable effects of checkerboard-free, mesh independence, crisp boundary, computational efficiency and conceptual simplicity. The proposed bilateral technique has a close relationship with the conventional domain filtering and range filtering. The proposed method is implemented in the framework of a power-law approach based on the optimality criteria and illustrated with 2D examples of minimum compliance design that has been extensively studied in the recent literature of topology optimization and its efficiency and accuracy are highlighted. Copyright © 2005 John Wiley & Sons, Ltd.

This paper addresses the problem of robust filtering in color images. We propose to improve the bilateral filter by using Ordered Weighted Averaging (OWA) filters. Bilateral filter de- creases amount of noise preserving edges when filtering. But the results could be strongly affected by the presence of outliers. In this paper, we define the filtering in the framework of fuzzy