Self-similar sketch
We introduce the self-similar sketch, a new method for the extraction of intermediate image features that combines three principles: detection of self-similarity structures, nonaccidental alignment, and instance-specific modelling. The method searches for self-similar image structures that form nona...
| Main Authors: | , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
Springer Nature
2012
|
| _version_ | 1824458711157964800 |
|---|---|
| author | Vedaldi, A Zisserman, A |
| author_facet | Vedaldi, A Zisserman, A |
| author_sort | Vedaldi, A |
| collection | OXFORD |
| description | We introduce the self-similar sketch, a new method for the extraction of intermediate image features that combines three principles: detection of self-similarity structures, nonaccidental alignment, and instance-specific modelling. The method searches for self-similar image structures that form nonaccidental patterns, for example collinear arrangements. We demonstrate a simple implementation of this idea where self-similar structures are found by looking for SIFT descriptors that map to the same visual words in image-specific vocabularies. This results in a visual word map which is searched for elongated connected components. Finally, segments are fitted to these connected components, extracting linear image structures beyond the ones that can be captured by conventional edge detectors, as the latter implicitly assume a specific appearance for the edges (steps). The resulting collection of segments constitutes a "sketch" of the image. This is applied to the task of estimating vanishing points, horizon, and zenith in standard benchmark data, obtaining state-of-the-art results. We also propose a new vanishing point estimation algorithm based on recently introduced techniques for the continuous-discrete optimisation of energies arising from model selection priors. © 2012 Springer-Verlag. |
| first_indexed | 2024-03-07T01:35:46Z |
| format | Conference item |
| id | oxford-uuid:9522d692-3f4f-4392-88c6-fdd8e165c344 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-02-19T04:30:14Z |
| publishDate | 2012 |
| publisher | Springer Nature |
| record_format | dspace |
| spelling | oxford-uuid:9522d692-3f4f-4392-88c6-fdd8e165c3442024-12-17T11:29:14ZSelf-similar sketchConference itemhttp://purl.org/coar/resource_type/c_5794uuid:9522d692-3f4f-4392-88c6-fdd8e165c344EnglishSymplectic Elements at OxfordSpringer Nature2012Vedaldi, AZisserman, AWe introduce the self-similar sketch, a new method for the extraction of intermediate image features that combines three principles: detection of self-similarity structures, nonaccidental alignment, and instance-specific modelling. The method searches for self-similar image structures that form nonaccidental patterns, for example collinear arrangements. We demonstrate a simple implementation of this idea where self-similar structures are found by looking for SIFT descriptors that map to the same visual words in image-specific vocabularies. This results in a visual word map which is searched for elongated connected components. Finally, segments are fitted to these connected components, extracting linear image structures beyond the ones that can be captured by conventional edge detectors, as the latter implicitly assume a specific appearance for the edges (steps). The resulting collection of segments constitutes a "sketch" of the image. This is applied to the task of estimating vanishing points, horizon, and zenith in standard benchmark data, obtaining state-of-the-art results. We also propose a new vanishing point estimation algorithm based on recently introduced techniques for the continuous-discrete optimisation of energies arising from model selection priors. © 2012 Springer-Verlag. |
| spellingShingle | Vedaldi, A Zisserman, A Self-similar sketch |
| title | Self-similar sketch |
| title_full | Self-similar sketch |
| title_fullStr | Self-similar sketch |
| title_full_unstemmed | Self-similar sketch |
| title_short | Self-similar sketch |
| title_sort | self similar sketch |
| work_keys_str_mv | AT vedaldia selfsimilarsketch AT zissermana selfsimilarsketch |