cvprlab

 
Home Research

DataFusionRealScratches

E-mail Print PDF

A Fusion-based Approach to Digital Movie Restoration: images affected by real scratches and related results



This page has been created in order to show the images affected by real scratches used for testing of the line scratch detection and removal algorithm presented in [1] and the related results.

Corrupted images

  1. Woman
Size: 352*288
Scratch width: 3
Source: [2]
Computed scratch masks and restored images
   
  2. Woman
Size: 320*120
Scratch width: 3
Source: [3]
Computed scratch masks and restored images
   
  3. Eye
Size: 220*220
Scratch width: 5
Source: [4] (detail)
Computed scratch masks and restored images
   
  4. Knight
Size: 256*256
Scratch width: 5
Source: [5]
Computed scratch masks and restored images
   
  5. Sitdwn
Size: 512*512
Scratch width: 7
Source: [5]
Computed scratch masks and restored images
   
  6. Star
Size: 720*576
Scratch width: 5
Source: [5]
Computed scratch masks and restored images
   

Computed scratch masks and restored images for corrupted image 1

  Detection alg. 1 Detection alg. 2 Detection alg. 3 Union fusion MC fusion
     
Scratch masks
    
  Removal algorithm 1 Removal algorithm 2 LSR algorithm
   
Restored images using Union mask
   
restored images using MC mask
   

Computed scratch masks and restored images for corrupted image 2

  Detection alg. 1 Detection alg. 2 Detection alg. 3 Union fusion MC fusion
     
Scratch masks
    
  Removal algorithm 1 Removal algorithm 2 LSR algorithm
   
Restored images using Union mask
   
restored images using MC mask
   

Computed scratch masks and restored images for corrupted image 3

  Detection alg. 1 Detection alg. 2 Detection alg. 3 Union fusion MC fusion
     
Scratch masks
    
  Removal algorithm 1 Removal algorithm 2 LSR algorithm
   
Restored images using Union mask
   
restored images using MC mask
   

Computed scratch masks and restored images for corrupted image 4

  Detection alg. 1 Detection alg. 2 Detection alg. 3 Union fusion MC fusion
     
Scratch masks
    
  Removal algorithm 1 Removal algorithm 2 LSR algorithm
   
Restored images using Union mask
   
restored images using MC mask
   

Computed scratch masks and restored images for corrupted image 5

  Detection alg. 1 Detection alg. 2 Detection alg. 3 Union fusion MC fusion
     
Scratch masks
    
  Removal algorithm 1 Removal algorithm 2 LSR algorithm
   
Restored images using Union mask
   
restored images using MC mask
   

Computed scratch masks and restored images for corrupted image 6

  Detection alg. 1 Detection alg. 2 Detection alg. 3 Union fusion MC fusion
     
Scratch masks
    
  Removal algorithm 1 Removal algorithm 2 LSR algorithm
   
Restored images using Union mask
   
restored images using MC mask
   

References

[1] L. Maddalena, A. Petrosino, G. Laccetti, A Fusion-based Approach to Digital Movie Restoration, Pattern Recognition, DOI 10.1016/j.patcog.2008.10.026, Vol. 42, no. 7, pagg. 1485-1495, 2009.

[2] DIAMANT Project (EC IST 1999 12078) - Digital Film Manipulation System

[3] Laboratoire d'Informatique et d'imagerie Industrielle (L3i), Université de La Rochelle - Sequences available at http://retouche.free.fr/download/results.htm

[4] Movie Animali che attraversano la strada, Italy, 2000

[5] A.C. Kokaram, Motion Picture Restoration: Digital Algorithms for Artefacts Suppression in Archived Film and Video, Springer-Verlag (1998).

 

Events & Announcements


 

Main events

VISMAC2018
IAPR Summer School on Machine and Vision Intelligence

CVPL2018
1° Convegno dell'Associazione Italiana per la ricerca in Computer Vision, Pattern recognition e machine Learning 

WILF2016
The Eleventh International Workshop on Fuzzy Logic and Applications
FLA2016
Workshop on Formal Language and Automata
 ICIAP 2013 
17th International Conference on Image Analysis and Processing September 11-13, 2013 Naples, Italy
PhD Studentship 2017 
in Machine Learning for Computer Vision