Abstract
This article describes an experimental procedure to analyze (and verify) the self-similarity concept in natural images and to explore the Gaussianity of groups of similar patches extracted from a single image. The self-similarity assumption means that most image patches of a sufficient size are repeated, of course not identically, but with small variations. The procedure proposed in this paper, and implemented in the accompanying online demo, permits to explore and visualize these clusters of similar patches in a given image. Thanks to it, a user can select a patch in an image, group all patches similar to it up to a translation, or to an isometry, apply PCA to the group, make visual tests about the Gaussianity of the set of patches, and finally apply EM to the set to see if it is a mixture of Gaussians.
| Original language | English |
|---|---|
| Pages (from-to) | 284-316 |
| Number of pages | 33 |
| Journal | Image Processing On Line |
| Volume | 11 |
| Early online date | 21 Sept 2021 |
| DOIs | |
| Publication status | Published - 2021 |
| Externally published | Yes |
Bibliographical note
Publisher Copyright:© 2021 IPOL & the authors.
Funding
The first author has been partially sponsored by MINECO/AEI/FEDER, UE projects TIN2017-85572-P, DPI2017-86372-C3-3-R, and by the Comunitat Autonoma de les Illes Balears through the Direcció General de Política Universitària i Recerca with funds from the Tourist Stay Tax Law ITS 2017-006 (PRD2018/26). The second author has been partly financed by Office of Naval research grant N00014-17-1-2552 and N00014-20-S-B001, DGA Astrid project “filmer la Terre” n ANR-17-ASTR-0013-01.
Keywords
- Eigenvectors
- Gaussian mixture
- Gaussianity
- Patches
- PCA
- Sparsity
Fingerprint
Dive into the research topics of 'Exploring patch similarity in an image'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver