Add to ORKGReranking is one of the commonly used methods to improve the initial ranking performance for content based object retrieval. In this paper, we propose a spatial consistency based selective reranking method to boost the performance of traditional reranking. After deriving the query's top results, we measure the spatial consistency degree of each query-result image pair via visual words spatial verification. Then the images which are most consistent to query image are selected as positive examples for current query images, and utilized in pseudo relevance feedback reranking. On the other hand, for queries whose top results are less consistent to the query images, they are excluded from reranking to avoid unnecessary performance deteriora...
As more and more images with user free tags are appear-ing on the Internet, image search reranking h...
Recently, various learning to rank approaches have been proposed in the information retrieval realm,...
Previous web image re-ranking approaches usually construct similarity measure on image level. Consid...
Reranking is one of the commonly used methods to improve the initial ranking performance for content...
Visual reranking has become a widely-accepted method to improve traditional text-based image search ...
International audienceWeb image search is inspired by text search techniques; it mainly relies on in...
Abstract — Search reranking is regarded as a common way to boost retrieval precision. The problem ne...
An image search reranking (ISR) technique aims at refining text-based search results by mining image...
Image search reranking methods usually fail to capture the user's intention when the query term is a...
Abstract: In this paper, a new method is proposed for object-based image re-trieval. The user sup...
Conference Name:21st ACM International Conference on Multimedia, MM 2013. Conference Address: Barcel...
This thesis focuses on efficient and effective object retrieval from an unlabelled collection of ima...
We propose a simple but effective re-ranking method for improving the results of object retrieval. O...
Query difficulty estimation predicts the performance of the search result of the given query. It is ...
The ranking models of existing image/video search engines are generally based on associated text whi...
As more and more images with user free tags are appear-ing on the Internet, image search reranking h...
Recently, various learning to rank approaches have been proposed in the information retrieval realm,...
Previous web image re-ranking approaches usually construct similarity measure on image level. Consid...
Reranking is one of the commonly used methods to improve the initial ranking performance for content...
Visual reranking has become a widely-accepted method to improve traditional text-based image search ...
International audienceWeb image search is inspired by text search techniques; it mainly relies on in...
Abstract — Search reranking is regarded as a common way to boost retrieval precision. The problem ne...
An image search reranking (ISR) technique aims at refining text-based search results by mining image...
Image search reranking methods usually fail to capture the user's intention when the query term is a...
Abstract: In this paper, a new method is proposed for object-based image re-trieval. The user sup...
Conference Name:21st ACM International Conference on Multimedia, MM 2013. Conference Address: Barcel...
This thesis focuses on efficient and effective object retrieval from an unlabelled collection of ima...
We propose a simple but effective re-ranking method for improving the results of object retrieval. O...
Query difficulty estimation predicts the performance of the search result of the given query. It is ...
The ranking models of existing image/video search engines are generally based on associated text whi...
As more and more images with user free tags are appear-ing on the Internet, image search reranking h...
Recently, various learning to rank approaches have been proposed in the information retrieval realm,...
Previous web image re-ranking approaches usually construct similarity measure on image level. Consid...