Transfer learning for object category detection
<p>Object category detection, the task of determining if one or more instances of a category are present in an image with their corresponding locations, is one of the fundamental problems of computer vision. The task is very challenging because of the large variations in imaged object appearan...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2014
|
| Subjects: |
| _version_ | 1797094689463074816 |
|---|---|
| author | Aytar, Y |
| author2 | Zisserman, A |
| author_facet | Zisserman, A Aytar, Y |
| author_sort | Aytar, Y |
| collection | OXFORD |
| description | <p>Object category detection, the task of determining if one or more instances of a category are present in an image with their corresponding locations, is one of the fundamental problems of computer vision. The task is very challenging because of the large variations in imaged object appearance, particularly due to the changes in viewpoint, illumination and intra-class variance. Although successful solutions exist for learning object category detectors, they require massive amounts of training data.</p> <p>Transfer learning builds upon previously acquired knowledge and thus reduces training requirements. The objective of this work is to develop and apply novel transfer learning techniques specific to the object category detection problem. This thesis proposes methods which not only address the challenges of performing transfer learning for object category detection such as finding relevant sources for transfer, handling aspect ratio mismatches and considering the geometric relations between the features; but also enable large scale object category detection by quickly learning from considerably fewer training samples and immediate evaluation of models on web scale data with the help of part-based indexing. Several novel transfer models are introduced such as: (a) rigid transfer for transferring knowledge between similar classes, (b) deformable transfer which tolerates small structural changes by deforming the source detector while performing the transfer, and (c) part level transfer particularly for the cases where full template transfer is not possible due to aspect ratio mismatches or not having adequately similar sources. Building upon the idea of using part-level transfer, instead of performing an exhaustive sliding window search, part-based indexing is proposed for efficient evaluation of templates enabling us to obtain immediate detection results in large scale image collections. Furthermore, easier and more robust optimization methods are developed with the help of feature maps defined between proposed transfer learning formulations and the “classical” SVM formulation.</p> |
| first_indexed | 2024-03-07T04:17:29Z |
| format | Thesis |
| id | oxford-uuid:c9e18ff9-df43-4f67-b8ac-28c3fdfa584b |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T04:17:29Z |
| publishDate | 2014 |
| record_format | dspace |
| spelling | oxford-uuid:c9e18ff9-df43-4f67-b8ac-28c3fdfa584b2022-03-27T07:03:11ZTransfer learning for object category detectionThesishttp://purl.org/coar/resource_type/c_db06uuid:c9e18ff9-df43-4f67-b8ac-28c3fdfa584bComputingEngineering & allied sciencesImage understandingEnglishOxford University Research Archive - Valet2014Aytar, YZisserman, A<p>Object category detection, the task of determining if one or more instances of a category are present in an image with their corresponding locations, is one of the fundamental problems of computer vision. The task is very challenging because of the large variations in imaged object appearance, particularly due to the changes in viewpoint, illumination and intra-class variance. Although successful solutions exist for learning object category detectors, they require massive amounts of training data.</p> <p>Transfer learning builds upon previously acquired knowledge and thus reduces training requirements. The objective of this work is to develop and apply novel transfer learning techniques specific to the object category detection problem. This thesis proposes methods which not only address the challenges of performing transfer learning for object category detection such as finding relevant sources for transfer, handling aspect ratio mismatches and considering the geometric relations between the features; but also enable large scale object category detection by quickly learning from considerably fewer training samples and immediate evaluation of models on web scale data with the help of part-based indexing. Several novel transfer models are introduced such as: (a) rigid transfer for transferring knowledge between similar classes, (b) deformable transfer which tolerates small structural changes by deforming the source detector while performing the transfer, and (c) part level transfer particularly for the cases where full template transfer is not possible due to aspect ratio mismatches or not having adequately similar sources. Building upon the idea of using part-level transfer, instead of performing an exhaustive sliding window search, part-based indexing is proposed for efficient evaluation of templates enabling us to obtain immediate detection results in large scale image collections. Furthermore, easier and more robust optimization methods are developed with the help of feature maps defined between proposed transfer learning formulations and the “classical” SVM formulation.</p> |
| spellingShingle | Computing Engineering & allied sciences Image understanding Aytar, Y Transfer learning for object category detection |
| title | Transfer learning for object category detection |
| title_full | Transfer learning for object category detection |
| title_fullStr | Transfer learning for object category detection |
| title_full_unstemmed | Transfer learning for object category detection |
| title_short | Transfer learning for object category detection |
| title_sort | transfer learning for object category detection |
| topic | Computing Engineering & allied sciences Image understanding |
| work_keys_str_mv | AT aytary transferlearningforobjectcategorydetection |