Non-Local Kalman: A Recursive Video Denoising Algorithm

Thibaud Ehret; Jean-Michel Morel; Pablo Arias

doi:10.1109/ICIP.2018.8451421

Non-Local Kalman: A Recursive Video Denoising Algorithm

Thibaud Ehret
, Jean-Michel Morel
, Pablo Arias

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

11 Citations (Scopus)

Abstract

In this article we propose a new recursive video denoising method with high performance. The method is recursive and uses only the current frame and the previous denoised one. It considers the video as a set of overlapping temporal patch trajectories. Following a Bayesian approach each trajectory is modeled as linear dynamic Gaussian model and denoised by a Kalman filter. To estimate its parameters, similar patches are grouped and their trajectories are considered as sharing the same model parameters. The filtering is mainly temporal; non-local spatial similarity is only used to estimate the parameters. This temporally causal method obtains results comparable (in terms of PSNR and SSIM) to state-of-the-art methods using several frames per frame denoised, but with a higher temporal consistency. © 2018 IEEE.

Original language	English
Title of host publication	2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings
Publisher	IEEE Computer Society
Pages	3204-3208
ISBN (Print)	9781479970612
DOIs	https://doi.org/10.1109/ICIP.2018.8451421
Publication status	Published - 29 Aug 2018
Externally published	Yes
Event	25th IEEE International Conference on Image Processing, ICIP 2018 - Athens, Greece Duration: 7 Oct 2018 → 10 Oct 2018

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
ISSN (Print)	1522-4880

Conference

Conference	25th IEEE International Conference on Image Processing, ICIP 2018
Place	Greece
City	Athens
Period	7/10/18 → 10/10/18

Bibliographical note

Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

Funding

Work supported by IDEX Paris-Saclay IDI 2016, ANR-11-IDEX-0003- 02, ONR grant N00014-17-1-2552, CNES MISS project, DGA Astrid ANR17-ASTR-0013-01, DGA ANR-16-DEFA-0004-01, and MENRT.

Research Keywords

Patch-based methods
Recursive filtering
Video denoising

Access Output

10.1109/ICIP.2018.8451421

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{7eebbb9c6f8743479a9c0c2b806a5519,

title = "Non-Local Kalman: A Recursive Video Denoising Algorithm",

abstract = "In this article we propose a new recursive video denoising method with high performance. The method is recursive and uses only the current frame and the previous denoised one. It considers the video as a set of overlapping temporal patch trajectories. Following a Bayesian approach each trajectory is modeled as linear dynamic Gaussian model and denoised by a Kalman filter. To estimate its parameters, similar patches are grouped and their trajectories are considered as sharing the same model parameters. The filtering is mainly temporal; non-local spatial similarity is only used to estimate the parameters. This temporally causal method obtains results comparable (in terms of PSNR and SSIM) to state-of-the-art methods using several frames per frame denoised, but with a higher temporal consistency. {\textcopyright} 2018 IEEE.",

keywords = "Patch-based methods, Recursive filtering, Video denoising",

author = "Thibaud Ehret and Jean-Michel Morel and Pablo Arias",

note = "Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected]. ; 25th IEEE International Conference on Image Processing, ICIP 2018 ; Conference date: 07-10-2018 Through 10-10-2018",

year = "2018",

month = aug,

day = "29",

doi = "10.1109/ICIP.2018.8451421",

language = "English",

isbn = "9781479970612",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "3204--3208",

booktitle = "2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings",

address = "United States",

}

Ehret, T, Morel, J-M & Arias, P 2018, Non-Local Kalman: A Recursive Video Denoising Algorithm. in 2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings., 8451421, Proceedings - International Conference on Image Processing, ICIP, IEEE Computer Society, pp. 3204-3208, 25th IEEE International Conference on Image Processing, ICIP 2018, Athens, Greece, 7/10/18. https://doi.org/10.1109/ICIP.2018.8451421

Non-Local Kalman: A Recursive Video Denoising Algorithm. / Ehret, Thibaud; Morel, Jean-Michel; Arias, Pablo.
2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings. IEEE Computer Society, 2018. p. 3204-3208 8451421 (Proceedings - International Conference on Image Processing, ICIP).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Non-Local Kalman

T2 - 25th IEEE International Conference on Image Processing, ICIP 2018

AU - Ehret, Thibaud

AU - Morel, Jean-Michel

AU - Arias, Pablo

N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

PY - 2018/8/29

Y1 - 2018/8/29

N2 - In this article we propose a new recursive video denoising method with high performance. The method is recursive and uses only the current frame and the previous denoised one. It considers the video as a set of overlapping temporal patch trajectories. Following a Bayesian approach each trajectory is modeled as linear dynamic Gaussian model and denoised by a Kalman filter. To estimate its parameters, similar patches are grouped and their trajectories are considered as sharing the same model parameters. The filtering is mainly temporal; non-local spatial similarity is only used to estimate the parameters. This temporally causal method obtains results comparable (in terms of PSNR and SSIM) to state-of-the-art methods using several frames per frame denoised, but with a higher temporal consistency. © 2018 IEEE.

AB - In this article we propose a new recursive video denoising method with high performance. The method is recursive and uses only the current frame and the previous denoised one. It considers the video as a set of overlapping temporal patch trajectories. Following a Bayesian approach each trajectory is modeled as linear dynamic Gaussian model and denoised by a Kalman filter. To estimate its parameters, similar patches are grouped and their trajectories are considered as sharing the same model parameters. The filtering is mainly temporal; non-local spatial similarity is only used to estimate the parameters. This temporally causal method obtains results comparable (in terms of PSNR and SSIM) to state-of-the-art methods using several frames per frame denoised, but with a higher temporal consistency. © 2018 IEEE.

KW - Patch-based methods

KW - Recursive filtering

KW - Video denoising

UR - https://www.scopus.com/pages/publications/85062900631

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85062900631&origin=recordpage

U2 - 10.1109/ICIP.2018.8451421

DO - 10.1109/ICIP.2018.8451421

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781479970612

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 3204

EP - 3208

BT - 2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings

PB - IEEE Computer Society

Y2 - 7 October 2018 through 10 October 2018

ER -

Non-Local Kalman: A Recursive Video Denoising Algorithm

Abstract

Publication series

Conference

Bibliographical note

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this