Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies

Arthur Renaudeau; François Lauze; Fabien Pierre; Jean François Aujol; Jean Denis Durou

doi:10.1007/978-3-030-22368-7_9

Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies

Arthur Renaudeau^*, François Lauze, Fabien Pierre, Jean François Aujol, Jean Denis Durou

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

2 Citations (Scopus)

Abstract

We propose a new video inpainting model for movies restoration application. Our model combines structural reconstruction with a diffusion-based method and textural reconstruction with a patch-based method. Both proposed energies (one for each method) are alternatively minimized in order to preserve the overall structure while adding textural refinement. While the structural reconstruction is obtained jointly with optical flow computation with several proximal approaches, the textural reconstruction is processed by a variational non-local approach. Preliminary results on different Middlebury frames show quality improvement in the reconstruction.

Original language	English
Title of host publication	Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings
Editors	Jan Lellmann, Jan Modersitzki, Martin Burger
Publisher	Springer
Publication date	2019
Pages	104-116
ISBN (Print)	9783030223670
DOIs	https://doi.org/10.1007/978-3-030-22368-7_9
Publication status	Published - 2019
Event	7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019 - Hofgeismar, Germany Duration: 30 Jun 2019 → 4 Jul 2019

Conference

Conference	7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019
Country/Territory	Germany
City	Hofgeismar
Period	30/06/2019 → 04/07/2019

Series	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11603 LNCS
ISSN	0302-9743

Bibliographical note

Publisher Copyright:
© 2019, Springer Nature Switzerland AG.

Access to Document

10.1007/978-3-030-22368-7_9

https://hal.archives-ouvertes.fr/hal-02433997/documentLicence: Unspecified

Cite this

Renaudeau, A., Lauze, F., Pierre, F., Aujol, J. F., & Durou, J. D. (2019). Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies. In J. Lellmann, J. Modersitzki, & M. Burger (Eds.), Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings (pp. 104-116). Springer. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 11603 LNCS https://doi.org/10.1007/978-3-030-22368-7_9

Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies. / Renaudeau, Arthur; Lauze, François; Pierre, Fabien; Aujol, Jean François; Durou, Jean Denis.

Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings. ed. / Jan Lellmann; Jan Modersitzki; Martin Burger. Springer, 2019. p. 104-116 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11603 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Renaudeau, A, Lauze, F, Pierre, F, Aujol, JF & Durou, JD 2019, Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies. in J Lellmann, J Modersitzki & M Burger (eds), Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings. Springer, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11603 LNCS, pp. 104-116, 7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019, Hofgeismar, Germany, 30/06/2019. https://doi.org/10.1007/978-3-030-22368-7_9

Renaudeau A, Lauze F, Pierre F, Aujol JF, Durou JD. Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies. In Lellmann J, Modersitzki J, Burger M, editors, Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings. Springer. 2019. p. 104-116. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11603 LNCS). https://doi.org/10.1007/978-3-030-22368-7_9

Renaudeau, Arthur ; Lauze, François ; Pierre, Fabien ; Aujol, Jean François ; Durou, Jean Denis. / Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies. Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings. editor / Jan Lellmann ; Jan Modersitzki ; Martin Burger. Springer, 2019. pp. 104-116 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11603 LNCS).

@inproceedings{bb9f2ab9cd994ddeb526d9fc8601e2c5,

title = "Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies",

abstract = "We propose a new video inpainting model for movies restoration application. Our model combines structural reconstruction with a diffusion-based method and textural reconstruction with a patch-based method. Both proposed energies (one for each method) are alternatively minimized in order to preserve the overall structure while adding textural refinement. While the structural reconstruction is obtained jointly with optical flow computation with several proximal approaches, the textural reconstruction is processed by a variational non-local approach. Preliminary results on different Middlebury frames show quality improvement in the reconstruction.",

author = "Arthur Renaudeau and Fran{\c c}ois Lauze and Fabien Pierre and Aujol, {Jean Fran{\c c}ois} and Durou, {Jean Denis}",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Nature Switzerland AG.; 7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019 ; Conference date: 30-06-2019 Through 04-07-2019",

year = "2019",

doi = "10.1007/978-3-030-22368-7_9",

language = "English",

isbn = "9783030223670",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "104--116",

editor = "Jan Lellmann and Jan Modersitzki and Martin Burger",

booktitle = "Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings",

address = "Switzerland",

}

TY - GEN

T1 - Alternate Structural-Textural Video Inpainting for Spot Defects Correction in Movies

AU - Renaudeau, Arthur

AU - Lauze, François

AU - Pierre, Fabien

AU - Aujol, Jean François

AU - Durou, Jean Denis

PY - 2019

Y1 - 2019

N2 - We propose a new video inpainting model for movies restoration application. Our model combines structural reconstruction with a diffusion-based method and textural reconstruction with a patch-based method. Both proposed energies (one for each method) are alternatively minimized in order to preserve the overall structure while adding textural refinement. While the structural reconstruction is obtained jointly with optical flow computation with several proximal approaches, the textural reconstruction is processed by a variational non-local approach. Preliminary results on different Middlebury frames show quality improvement in the reconstruction.

AB - We propose a new video inpainting model for movies restoration application. Our model combines structural reconstruction with a diffusion-based method and textural reconstruction with a patch-based method. Both proposed energies (one for each method) are alternatively minimized in order to preserve the overall structure while adding textural refinement. While the structural reconstruction is obtained jointly with optical flow computation with several proximal approaches, the textural reconstruction is processed by a variational non-local approach. Preliminary results on different Middlebury frames show quality improvement in the reconstruction.

UR - http://www.scopus.com/inward/record.url?scp=85068464312&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-22368-7_9

DO - 10.1007/978-3-030-22368-7_9

M3 - Article in proceedings

AN - SCOPUS:85068464312

SN - 9783030223670

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 104

EP - 116

BT - Scale Space and Variational Methods in Computer Vision - 7th International Conference, SSVM 2019, Proceedings

A2 - Lellmann, Jan

A2 - Modersitzki, Jan

A2 - Burger, Martin

PB - Springer

T2 - 7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019

Y2 - 30 June 2019 through 4 July 2019

ER -