Self-adaptation in Image and Video Retrieval - Multimedia Database Retrieval: Technology and Applications - page 58

Chapter 3
Self-adaptation in Image and Video Retrieval

Abstract This chapter explores the automatic methods for implementing
pseudo-relevance feedback for retrieval of images and videos. The automation
is based on dynamic self-organization, the self-organizing tree map that is capable
of identification of relevance in place of human users. The automation process leads
to the avoidance of errors in excessive human involvement, and enlarging the size of
training set, as compared to traditional relevance feedback. The automatic retrieval
system applies for image retrieval in compressed domains (i.e., JPEG and wavelet
based coders). In addition, the system incorporates knowledge-based learning to
acquire a suitable weighting scheme for unsupervised relevance identification. In the
video domain, the pseudo-relevance feedback is implemented by an adaptive cosine
network than enhances retrieval accuracy through the network's forward-backward
signal propagation, without user input.

3.1 Introduction

In order to handle the large volumes of multimedia information that are becom-
ing readily accessible in the consumer and the industrial world, some level of
automation is desirable. Automation requires intelligence systems, to formulate
its own models of the data in question with little or no user intervention. The
system is able to make decisions about what information is actually important
and what is not. In effect, like a human user, the system must be able to discover
characteristic properties of data in some appropriate manner, without a teacher. This
process is known as unsupervised learning , and in this chapter we explore its use in
performing relevance identification in place of human users in relevance feedback
based multimedia retrieval.
This chapter introduces self-adaptation methods for the automation of adaptive
retrieval systems in image and video database applications. This aims to achieve the
following advantages in overcoming the difficulties faced in traditional relevance
feedback,

￿ Avoiding errors caused by excessive human involvement in relevance feedback

loops, thus, offering a more-user friendly environment.