Applications of Video Segmentation - Video Segmentation and Its Applications - page 145

Chapter 6
Applications of Video Segmentation

E. Izquierdo and K. Vaiapury

Abstract Segmentation is one of the important computer vision processes that
is used in many practical applications such as medical imaging, computer-guided
surgery, machine vision, object recognition, surveillance, content-based browsing,
augmented reality applications, etc. . The knowledge to ascertain plausible seg-
mentation applications and corresponding algorithmic techniques is necessary to
simplify the video representation into a more meaningful and easier form to ana-
lyze. This is because expected segmentation quality for a given application depends
on the level of granularity and the requirement that is related to shape precision and
temporal coherence of the objects.

6.1 Introduction

With the rapid growth of video data, management, access, and retrieval of desired
information from humongous video library is becoming a headachy experience for
users. Segmentation is one of the important computer vision processes that is used in
many practical applications such as medical imaging, computer-guided surgery, ma-
chine vision, object recognition, surveillance, content-based browsing, augmented
reality applications, etc . The knowledge to ascertain plausible segmentation appli-
cations and corresponding algorithmic techniques is necessary to simplify the video
representation into a more meaningful and easier form to analyze. In fact, expected
segmentation quality for a given application depends on the level of granularity and
the requirement that is related to shape precision and temporal coherence of the
objects. In this chapter, we discuss key applications of video segmentation. Video
Segmentation refers to the process of splitting videos into homogenous spatial tem-
poral segments meaningful from a semantic point of view. Video is a sequence of
frames that have a high degree of temporal correlation among them [ 21 ]. Each frame
is an image in 2D spatial plane. Though the underlying segmentation process is the

E. Izquierdo ( )

Department of Electronic Engineering, Queen Mary, University of London, London, UK
e-mail: [e-mail]