All about VLAD
The objective of this paper is large scale object instance retrieval, given a query image. A starting point of such systems is feature detection and description, for example using SIFT. The focus of this paper, however, is towards very large scale retrieval where, due to storage requirements, very c...
| Main Authors: | , |
|---|---|
| Format: | Conference item |
| Published: |
IEEE
2013
|
| _version_ | 1824458862237843456 |
|---|---|
| author | Arandjelović, R Zisserman, A |
| author_facet | Arandjelović, R Zisserman, A |
| author_sort | Arandjelović, R |
| collection | OXFORD |
| description | The objective of this paper is large scale object instance retrieval, given a query image. A starting point of such systems is feature detection and description, for example using SIFT. The focus of this paper, however, is towards very large scale retrieval where, due to storage requirements, very compact image descriptors are required and no information about the original SIFT descriptors can be accessed directly at run time. We start from VLAD, the state-of-the art compact descriptor introduced by Jegou et al. for this purpose, and make three novel contributions: first, we show that a simple change to the normalization method significantly improves retrieval performance; second, we show that vocabulary adaptation can substantially alleviate problems caused when images are added to the dataset after initial vocabulary learning. These two methods set a new state-of-the-art over all benchmarks investigated here for both mid-dimensional (20k-D to 30k-D) and small (128-D) descriptors. Our third contribution is a multiple spatial VLAD representation, MultiVLAD, that allows the retrieval and localization of objects that only extend over a small part of an image (again without requiring use of the original image SIFT descriptors). |
| first_indexed | 2024-03-06T19:09:20Z |
| format | Conference item |
| id | oxford-uuid:163cd743-4b36-46ab-bd00-7c1079373802 |
| institution | University of Oxford |
| last_indexed | 2025-02-19T04:32:38Z |
| publishDate | 2013 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:163cd743-4b36-46ab-bd00-7c10793738022025-01-16T10:55:39ZAll about VLADConference itemhttp://purl.org/coar/resource_type/c_5794uuid:163cd743-4b36-46ab-bd00-7c1079373802Symplectic Elements at OxfordIEEE2013Arandjelović, RZisserman, AThe objective of this paper is large scale object instance retrieval, given a query image. A starting point of such systems is feature detection and description, for example using SIFT. The focus of this paper, however, is towards very large scale retrieval where, due to storage requirements, very compact image descriptors are required and no information about the original SIFT descriptors can be accessed directly at run time. We start from VLAD, the state-of-the art compact descriptor introduced by Jegou et al. for this purpose, and make three novel contributions: first, we show that a simple change to the normalization method significantly improves retrieval performance; second, we show that vocabulary adaptation can substantially alleviate problems caused when images are added to the dataset after initial vocabulary learning. These two methods set a new state-of-the-art over all benchmarks investigated here for both mid-dimensional (20k-D to 30k-D) and small (128-D) descriptors. Our third contribution is a multiple spatial VLAD representation, MultiVLAD, that allows the retrieval and localization of objects that only extend over a small part of an image (again without requiring use of the original image SIFT descriptors). |
| spellingShingle | Arandjelović, R Zisserman, A All about VLAD |
| title | All about VLAD |
| title_full | All about VLAD |
| title_fullStr | All about VLAD |
| title_full_unstemmed | All about VLAD |
| title_short | All about VLAD |
| title_sort | all about vlad |
| work_keys_str_mv | AT arandjelovicr allaboutvlad AT zissermana allaboutvlad |