Non-Blind Deconvolution of Point Cloud Attributes in Graph Spectral Domain

Kaoru YAMAMOTO; Masaki ONUKI; Yuichi TANAKA

doi:10.1587/transfun.E100.A.1751

Non-Blind Deconvolution of Point Cloud Attributes in Graph Spectral Domain

Kaoru YAMAMOTO, Masaki ONUKI, Yuichi TANAKA

Full Text Views

0

Cite this

Summary :

We propose a non-blind deconvolution algorithm of point cloud attributes inspired by multi-Wiener SURE-LET deconvolution for images. The image reconstructed by the SURE-LET approach is expressed as a linear combination of multiple filtered images where the filters are defined on the frequency domain. The coefficients of the linear combination are calculated so that the estimate of mean squared error between the original and restored images is minimized. Although the approach is very effective, it is only applicable to images. Recently we have to handle signals on irregular grids, e.g., texture data on 3D models, which are often blurred due to diffusion or motions of objects. However, we cannot utilize image processing-based approaches straightforwardly since these high-dimensional signals cannot be transformed into their frequency domain. To overcome the problem, we use graph signal processing (GSP) for deblurring the complex-structured data. That is, the SURE-LET approach is redefined on GSP, where the Wiener-like filtering is followed by the subband decomposition with an analysis graph filter bank, and then thresholding for each subband is performed. In the experiments, the proposed method is applied to blurred textures on 3D models and synthetic sparse data. The experimental results show clearly deblurred signals with SNR improvements.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E100-A No.9 pp.1751-1759

Publication Date: 2017/09/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E100.A.1751

Type of Manuscript: Special Section PAPER (Special Section on Signal Processing on Irregular Sampling Grids)

Category

Authors

Kaoru YAMAMOTO
  Tokyo University of Agriculture and Technology
Masaki ONUKI
  Tokyo University of Agriculture and Technology
Yuichi TANAKA
  Tokyo University of Agriculture and Technology,Japan Science and Technology Agency

Keyword

graph signal processing, point cloud, high-dimensional signal, SURE, 3D model

Cite this

Copy

Kaoru YAMAMOTO, Masaki ONUKI, Yuichi TANAKA, "Non-Blind Deconvolution of Point Cloud Attributes in Graph Spectral Domain" in IEICE TRANSACTIONS on Fundamentals, vol. E100-A, no. 9, pp. 1751-1759, September 2017, doi: 10.1587/transfun.E100.A.1751.
Abstract: We propose a non-blind deconvolution algorithm of point cloud attributes inspired by multi-Wiener SURE-LET deconvolution for images. The image reconstructed by the SURE-LET approach is expressed as a linear combination of multiple filtered images where the filters are defined on the frequency domain. The coefficients of the linear combination are calculated so that the estimate of mean squared error between the original and restored images is minimized. Although the approach is very effective, it is only applicable to images. Recently we have to handle signals on irregular grids, e.g., texture data on 3D models, which are often blurred due to diffusion or motions of objects. However, we cannot utilize image processing-based approaches straightforwardly since these high-dimensional signals cannot be transformed into their frequency domain. To overcome the problem, we use graph signal processing (GSP) for deblurring the complex-structured data. That is, the SURE-LET approach is redefined on GSP, where the Wiener-like filtering is followed by the subband decomposition with an analysis graph filter bank, and then thresholding for each subband is performed. In the experiments, the proposed method is applied to blurred textures on 3D models and synthetic sparse data. The experimental results show clearly deblurred signals with SNR improvements.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E100.A.1751/_p

Copy

@ARTICLE{e100-a_9_1751,
author={Kaoru YAMAMOTO, Masaki ONUKI, Yuichi TANAKA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Non-Blind Deconvolution of Point Cloud Attributes in Graph Spectral Domain},
year={2017},
volume={E100-A},
number={9},
pages={1751-1759},
abstract={We propose a non-blind deconvolution algorithm of point cloud attributes inspired by multi-Wiener SURE-LET deconvolution for images. The image reconstructed by the SURE-LET approach is expressed as a linear combination of multiple filtered images where the filters are defined on the frequency domain. The coefficients of the linear combination are calculated so that the estimate of mean squared error between the original and restored images is minimized. Although the approach is very effective, it is only applicable to images. Recently we have to handle signals on irregular grids, e.g., texture data on 3D models, which are often blurred due to diffusion or motions of objects. However, we cannot utilize image processing-based approaches straightforwardly since these high-dimensional signals cannot be transformed into their frequency domain. To overcome the problem, we use graph signal processing (GSP) for deblurring the complex-structured data. That is, the SURE-LET approach is redefined on GSP, where the Wiener-like filtering is followed by the subband decomposition with an analysis graph filter bank, and then thresholding for each subband is performed. In the experiments, the proposed method is applied to blurred textures on 3D models and synthetic sparse data. The experimental results show clearly deblurred signals with SNR improvements.},
keywords={},
doi={10.1587/transfun.E100.A.1751},
ISSN={1745-1337},
month={September},}

Copy

TY - JOUR
TI - Non-Blind Deconvolution of Point Cloud Attributes in Graph Spectral Domain
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1751
EP - 1759
AU - Kaoru YAMAMOTO
AU - Masaki ONUKI
AU - Yuichi TANAKA
PY - 2017
DO - 10.1587/transfun.E100.A.1751
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E100-A
IS - 9
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - September 2017
AB - We propose a non-blind deconvolution algorithm of point cloud attributes inspired by multi-Wiener SURE-LET deconvolution for images. The image reconstructed by the SURE-LET approach is expressed as a linear combination of multiple filtered images where the filters are defined on the frequency domain. The coefficients of the linear combination are calculated so that the estimate of mean squared error between the original and restored images is minimized. Although the approach is very effective, it is only applicable to images. Recently we have to handle signals on irregular grids, e.g., texture data on 3D models, which are often blurred due to diffusion or motions of objects. However, we cannot utilize image processing-based approaches straightforwardly since these high-dimensional signals cannot be transformed into their frequency domain. To overcome the problem, we use graph signal processing (GSP) for deblurring the complex-structured data. That is, the SURE-LET approach is redefined on GSP, where the Wiener-like filtering is followed by the subband decomposition with an analysis graph filter bank, and then thresholding for each subband is performed. In the experiments, the proposed method is applied to blurred textures on 3D models and synthetic sparse data. The experimental results show clearly deblurred signals with SNR improvements.
ER -