Skip to Main content Skip to Navigation
Theses

Détection des défauts et inpainting vidéo pour la restauration de films

Arthur Renaudeau 1
1 IRIT-REVA - Real Expression Artificial Life
IRIT - Institut de recherche en informatique de Toulouse
Abstract : Cinema was invented more than a century ago. Many films have been shot, following the technological evolutions of each era, from the Lumière brothers’ cinematograph to 3D. The medium of these films has also evolved over time, from silver salts on film to hard disk for digital. With time, the supports of silver films having undergone chemical degradations, it becomes essential to restore them in order to be able to enjoy their content in good conditions. This is a major heritage conservation issue, but raises ethical questions about how far a film should be restored. In this context, and in collaboration with the Cinémathèque de Toulouse, we have developed an automatic restoration process for old movies that consists of two steps. The first step is to detect the defects present in an image. The second step consists in filling the areas thus detected by video inpainting. Taking into account the temporal information contained in the images adjacent to the degraded image is an essential aspect of both these steps. The detection of defects such as spots, dust or other defects is performed by automatic learning on deep neural networks. In particular, a U-Net receiving three successive images as input can detect temporal inconsistencies characteristic of defects. The output of this network is compared to a defect mask created from the restored version of the central image obtained with specialized software manipulated by an expert from the Cinémathèque de Toulouse. Finally, the filling of the damaged areas is carried out by alternating the reconstruction of the structure and the reconstruction of the texture of the image to be restored, both of which carry out the search for an optimum using variational methods.
Complete list of metadata

https://hal.archives-ouvertes.fr/tel-03186687
Contributor : Arthur Renaudeau <>
Submitted on : Wednesday, March 31, 2021 - 11:48:48 AM
Last modification on : Tuesday, May 4, 2021 - 4:07:42 PM

File

ARTHUR_Manuscrit_2020_final.pd...
Files produced by the author(s)

Identifiers

  • HAL Id : tel-03186687, version 1

Citation

Arthur Renaudeau. Détection des défauts et inpainting vidéo pour la restauration de films. Traitement des images [eess.IV]. Toulouse INP, 2021. Français. ⟨tel-03186687⟩

Share

Metrics

Record views

28

Files downloads

14