Abstract
Nowadays, web-scale image search engines (e.g. Google Image Search, Microsoft Live Image Search) rely almost purely on surrounding text features. This leads to the ambiguous and noisy results. We propose an adaptive visual similarity to re-rank the text based search results. A query image is first categorized as one of several predefined intention categories, and a specific similarity measure has been used inside each of category to combine the image features for re-ranking based on the query image. The Extensive experiments demonstrate that using this algorithm to filter output of Google Image Search and Microsoft Live Image Search is a practical and effective way to dramatically improve the user’s experience.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have