Learning object categories from Google’s image search
Current approaches to object category recognition require datasets of training images to be manually prepared, with varying degrees of supervision. We present an approach that can learn an object category from just its name, by utilizing the raw output of image search engines available on the Intern...
| Main Authors: | , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
IEEE
2005
|
| _version_ | 1824458866303172608 |
|---|---|
| author | Fergus, R Fei-Fei, L Perona, P Zisserman, A |
| author_facet | Fergus, R Fei-Fei, L Perona, P Zisserman, A |
| author_sort | Fergus, R |
| collection | OXFORD |
| description | Current approaches to object category recognition require datasets of training images to be manually prepared, with varying degrees of supervision. We present an approach that can learn an object category from just its name, by utilizing the raw output of image search engines available on the Internet. We develop a new model, TSI-pLSA, which extends pLSA (as applied to visual words) to include spatial information in a translation and scale invariant manner. Our approach can handle the high intra-class variability and large proportion of unrelated images returned by search engines. We evaluate tire models on standard test sets, showing performance competitive with existing methods trained on hand prepared datasets. |
| first_indexed | 2025-02-19T04:32:42Z |
| format | Conference item |
| id | oxford-uuid:20c5bc69-71d6-4e7e-bcb9-e8590b99b32e |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-02-19T04:32:42Z |
| publishDate | 2005 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:20c5bc69-71d6-4e7e-bcb9-e8590b99b32e2025-01-15T15:02:03ZLearning object categories from Google’s image searchConference itemhttp://purl.org/coar/resource_type/c_5794uuid:20c5bc69-71d6-4e7e-bcb9-e8590b99b32eEnglishSymplectic ElementsIEEE2005Fergus, RFei-Fei, LPerona, PZisserman, ACurrent approaches to object category recognition require datasets of training images to be manually prepared, with varying degrees of supervision. We present an approach that can learn an object category from just its name, by utilizing the raw output of image search engines available on the Internet. We develop a new model, TSI-pLSA, which extends pLSA (as applied to visual words) to include spatial information in a translation and scale invariant manner. Our approach can handle the high intra-class variability and large proportion of unrelated images returned by search engines. We evaluate tire models on standard test sets, showing performance competitive with existing methods trained on hand prepared datasets. |
| spellingShingle | Fergus, R Fei-Fei, L Perona, P Zisserman, A Learning object categories from Google’s image search |
| title | Learning object categories from Google’s image search |
| title_full | Learning object categories from Google’s image search |
| title_fullStr | Learning object categories from Google’s image search |
| title_full_unstemmed | Learning object categories from Google’s image search |
| title_short | Learning object categories from Google’s image search |
| title_sort | learning object categories from google s image search |
| work_keys_str_mv | AT fergusr learningobjectcategoriesfromgooglesimagesearch AT feifeil learningobjectcategoriesfromgooglesimagesearch AT peronap learningobjectcategoriesfromgooglesimagesearch AT zissermana learningobjectcategoriesfromgooglesimagesearch |