An Attribute-Assisted Reranking Model for Web Image
Search
ABSTRACT:
Image search reranking is an effective approach to refine the text-based image
search result. Most existing reranking approaches are based on low-level visual
features. In this paper, we propose to exploit semantic attributes for image search
reranking. Based on the classifiers for all the predefined attributes, each image is
represented by an attribute feature consisting of the responses from these
classifiers. A hypergraph is then used to model the relationship between images by
integrating low-level visual features and attribute features. Hypergraph ranking is
then performed to order the images. Its basic principle is that visually similar
images should have similar ranking scores. In this paper, we propose a visualattribute joint hypergraph learning approach to simultaneously explore two
information sources. A hypergraph is constructed to model the relationship of all
images.We conduct experiments on more than 1,000 queries in MSRA-MMV2.0
data set. The experimental results demonstrate the effectiveness of our approach.
EXISTING SYSTEM:
 Many image search engines such as Google and Bing have relied on
matching textual information of the images against queries given by users.
However, text-based image retrieval suffers from essential difficulties that
are caused mainly by the incapability of the associated text to appropriately
describe the image content.
 The existing visual reranking methods can be typically categorized into three
categories as the clustering based, classification based and graph based
methods.
 The clustering based reranking methods stem from the key observation that a
wealth of visual characteristics can be shared by relevant images.
 In the classification based methods, visual reranking is formulated as binary
classification problem aiming to identify whether each search result is
relevant or not.
 Graph based methods have been proposed recently and received increasing
attention as demonstrated to be effective. The multimedia entities in top
ranks and their visual relationship can be represented as a collection of
nodes and edges
DISADVANTAGES OF EXISTING SYSTEM:
 In the classification based methods, visual reranking is formulated as binary
classification problem aiming to identify whether each search result is
relevant or not.
 The framework casts the reranking problem as random walk on an affinity
graph and reorders images according to the visual similarities.
PROPOSED SYSTEM:
 We propose a new attribute-assisted reranking method based on hypergraph
learning. We first train several classifiers for all the pre-defined attributes
and each image is represented by attribute feature consisting of the responses
from these classifiers.
 We improve the hypergraph learning method approach by adding a
regularizer on the hyperedge weights which performs an implicit selection
on the semantic attributes.
 This paper serves as a first attempt to include the attributes in reranking
framework. We observe that semantic attributes are expected to narrow
down the semantic gap between low-level visual features and high level
semantic meanings.
ADVANTAGES OF PROPOSED SYSTEM:
 We propose a novel attribute-assisted retrieval model for reranking images.
Based on the classifiers for all the predefined attributes.
 We perform hypergraph ranking to re-order the images, which is also
constructed to model the relationship of all images.
 Our proposed iterative regularization framework could further explore the
semantic similarity between images by aggregating their local
 Compared with the previous method, a hypergraph is reconstructed to model
the relationship of all the images, in which each vertex denotes an image and
a hyperedge represents an attribute and a hyperedge connects to multiple
vertices.
SYSTEM ARCHITECTURE:
SYSTEM REQUIREMENTS:
HARDWARE REQUIREMENTS:
 System
:
Pentium IV 2.4 GHz.
 Hard Disk
:
40 GB.
 Floppy Drive
:
1.44 Mb.
 Monitor
:
15 VGA Colour.
 Mouse
:
Logitech.
 Ram
:
512 Mb.
SOFTWARE REQUIREMENTS:
 Operating system :
Windows XP/7.
 Coding Language :
JAVA/J2EE
 IDE
:
Netbeans 7.4
 Database
:
MYSQL
REFERENCE:
Junjie Cai, Zheng-Jun Zha, Member, IEEE, Meng Wang, Shiliang Zhang, and Qi
Tian, Senior Member, IEEE, “An Attribute-Assisted Reranking Model for Web
Image Search”, IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL.
24, NO. 1, JANUARY 2015.