Image Re-Ranking Using Query-Specific Semantic Signature with Result Analysis


Image Re-Ranking Using Query-Specific Semantic Signature with Result Analysis

Ms. Tejashree Kumar Shinde, Prof. Prakash. B. Dhainje, Dr.Deshmukh Pradeep K

Ms. Tejashree Kumar Shinde, Prof. Prakash. B. Dhainje, Dr.Deshmukh Pradeep K "Image Re-Ranking Using Query-Specific Semantic Signature with Result Analysis" Published in International Journal of Trend in Research and Development (IJTRD), ISSN: 2394-9333, Volume-2 | Issue-4 , August 2015, URL: http://www.ijtrd.com/papers/IJTRD19.pdf

For searching images, Image Search engines mostly use keywords and they rely on surrounding text. Indistinctness of query images is hard to describe accurately by using keywords. Eg. If Apple is query keyword then categories can be ”apple laptop” , “red apple”, etc. Without online training low level features may not well co-relate with high level semantic meaning is one challenge. Some Low-level features are sometimes incompatible with visual observation. To get semantic signature the visual and textual features of images are then projected into their related semantic spaces. In online stage images are re-ranked by comparing semantic signature obtained from semantic space obtained from query keywords. By just 20 – 30 concepts Semantic space of a query keyword can be described these are referred as “reference classes”.

K-Means Algorithm, Semantic Signatures, Canny Edge Detection, Re-ranking, framework


Volume-2 | Issue-4 , August 2015

2394-9333

IJTRD19
pompy wtryskowe|cheap huarache shoes| cheap jordans|cheap jordans|cheap air max| cheap sneaker cheap nfl jerseys|cheap air jordanscheap jordan shoes
cheap wholesale jordans