05-07-2012, 04:12 PM
Active Re-ranking for Web Image Search
Active Re-ranking for Web Image Search.pptx (Size: 976.31 KB / Downloads: 76)
Abstract
Image search re-ranking methods usually fails to capture the user’s intention when the query term is ambiguous. Therefore, re ranking with user interactions, or active re ranking, is highly demanded to effectively improve the search performance. The essential issue in active re ranking is how to target the user’s intention. To complete this goal, this paper presents a structural information based sample selection strategy to reduce the user’s labeling efforts. Furthermore, to localize the user’s intention in the visual feature space, a novel local-global discriminative dimension reduction algorithm is proposed. In this algorithm, a sub manifold is learned by transferring the local geometry and the discriminative information from the labeled images to the whole (global) image database.
Existing system:
Image search re-ranking methods usually fail to capture the user’s intention when the query term is ambiguous.
2 . Text-based search techniques have shown their effectiveness in the document search, they are problematic when applied to the image search.
Disadvantages:
User intention is not considering in the already existing system.
Re-ranking methods usually fail to capture the user’s intention when the query term is ambiguous.
Text-based search techniques are problematic when applied to the image search because of the mismatching between images and their associated textual information.
Textual information is insufficient to represent the semantic content of the images.
Advantages:
Re-ranking with user interactions, or active reranking is introduced in this system.
Collecting labeling information from users to obtain the specified semantic space.
localizing the visual characteristics of the user’s intention in this specific semantic space
A new structural information (SInfo) based strategy is proposed to actively select the most informative query images.
To localize the visual characteristics of the user’s intention, we propose a novel local-global discriminative (LGD) dimension reduction algorithm.