An Approach to Re-Rank Retrieved Images - Image Search Results with Visual Similarity Presented by Veningston .K M.E student, Dept of CSE, Anna University – Coimbatore, India. Guided by Mr. M. Newlin Rajkumar Lecturer, Dept of CSE, Anna University – Coimbatore, India.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
An Approach to Re-Rank Retrieved Images - Image Search Results with Visual
Similarity
Presented by
Veningston .KM.E student, Dept of CSE,Anna University – Coimbatore,
India.
Guided by
Mr. M. Newlin RajkumarLecturer, Dept of CSE,
Anna University – Coimbatore, India.
04/11/23 Department of Computer Science and Engineering
2
Objective
To address a ranking problem in web image retrieval
System to re-rank images returned by image search engine
Re-ranking images by incorporating,visual aspects visual similarity
04/11/23 Department of Computer Science and Engineering
3
Introduction
Image Retrieval System Comparative study on text & image based search Interest point extraction - visual content of images Re-ranking the results of text based systems
using visual information Finding the largest set of most similar images Rearranging images based on the similarity
04/11/23 Department of Computer Science and Engineering
4
Why image Re-ranking?
To maximize relevancy of image results To achieve diversity of image results
04/11/23 Department of Computer Science and Engineering
5
Proposed Scheme
Goal Retrieve image results that are relevant Finding common features among images
Overview Interest points on the images are extracted Similarity of each pair of images are computed Generate graph model Apply page ranking
04/11/23 Department of Computer Science and Engineering
6
Finding common features among images
Similarity measurement to handle potential rotation, scale and perspective transformations.
04/11/23 Department of Computer Science and Engineering
7
Scale Invariant Feature Transform Image matching and features to similarity
04/11/23 Department of Computer Science and Engineering
8
Image similarity
Given two images u and v, Corresponding descriptor vector, Du = (d1
u, d2
u, ...dmu ) and Dv = (d1
v, d2v, ...dn
v ), Define the similarity between two images
simply as the number of interest points shared between two images divided by their average number of interest points.
04/11/23 Department of Computer Science and Engineering
9
Graph model
Given the visual similarities of the images to be ranked. Treat images as web documents Treat similarities as visual hyperlinks Estimate the probability of images being visited by
a user // using page rank Images with more estimated visits are ranked
higher
04/11/23 Department of Computer Science and Engineering
10
Similarity to Centrality
Centrality – importance of images Given a graph with vertices and a set of
weighted edges, define and measure the “importance” of each of the vertices
Vertices = images Egde weights = similarity
A vertex closer to an important vertex should rank higher than others
04/11/23 Department of Computer Science and Engineering
11
Similarity to Centrality Ranking scores correspond to probability of
arriving in each vertex by traversing through the graph
Adjacency matrix for unweighted graph
Matrix constructed from the weights of the edges in the
graph
Decision to take a particular path defined by
weighted edges
04/11/23 Department of Computer Science and Engineering
12
Page Ranking (PR)
Determines the order of search results
Method of measuring a page’s importance
Results are based on this priority order
AA BB
CC
Search result Priority
B 3
A 2
C 1
D 1
E 1
DDEE
04/11/23 Department of Computer Science and Engineering
13
Overall scheme
Computationally infeasible to compute similarities for all images indexed by search engine
Pre-cluster web images based on metadata Define the similarity of images Given a query, extract top - N results
returned, create graph of visual similarity on the N images
Compute image rank only on this subset
04/11/23 Department of Computer Science and Engineering
04/11/23 Department of Computer Science and Engineering
15
Merits
Minimizing irrelevant images Selecting small set of images Computational cost
04/11/23 Department of Computer Science and Engineering
16
Conclusion
Ability to reduce the number of irrelevant images
A tiny set of important images can be selected from a very large set of candidates
04/11/23 Department of Computer Science and Engineering
17
Future work
Extensions of this technique to a query driven feature selection.
04/11/23 Department of Computer Science and Engineering
18
References
[1] Yushi Jing, Shumeet Baluja. VisualRank: Applying PageRank to Large-Scale Image Search. IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol 30, No. 11:1887–1890, 2008.
[2] R. Datta, D. Joshi, J. Li, and J. Wang. Image Retrieval: Ideas, Influences, and Trends of the New Age. ACM Computing Surveys, vol. 40, no. 2, 2008.
[3] Yushi Jing, Shumeet Baluja. PageRank for Product Image Search. WWW 2008, April 21–25, 2008, Beijing, China. ACM International conf. 2008.
[4] K. Mikolajczyk and C. Schmid. A performance evaluation of local descriptors. IEEE Transaction on Pattern Analysis and Machine Intelligence, 27(10):1615–1630, 2005.
[5] S. Belongie, J. Malik, and J. Puzicha. Shape matching and object recognition using shape contexts. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 24(24):509–522, 2002.
04/11/23 Department of Computer Science and Engineering