Game Development Reference
In-Depth Information
Chapter 8
Scene Video Coding
Scene videos are defined as the kinds of video captured from a scene for a long while.
Specifically, surveillance video is a typical kind of scene video, and this chapter
will introduce some efficient coding tools developed for surveillance video in AVS,
which also work well for other common scene video, e.g., conference video. This
chapter consists of five parts. The first part introduces the background redundancy
features of surveillance video. The second part gives a brief overview of model-based
coding. The third part details the background picture model-based coding method,
and typical background models are surveyed. In the fourth part, background picture-
based surveillance coding methods in AVS and AVS2 are provided. The last part
summarizes this chapter.
8.1 The Redundancy Features of Surveillance Video
The past 1970s have seen an explosion around the world in the use of video sur-
veillance in many places from law enforcement to traffic control, public security,
and in-home nanny cam monitoring. Nowadays, different numbers of surveillance
cameras are connected to recording and display devices via transmission networks,
and constitute the so-called video surveillance systems at different scales.
According to a recent report by IDC (
2012
), by 2020, as many as 5,800 exabytes
of surveillance videos will be saved, transmitted, and analyzed. Traditionally, the
industry often employs the generic video coding standards such as MPEG.4 and
AVC/H.264 to compress surveillance videos, which were originally designed for
broadcasting videos. If we still follow this technology road map, in the future sev-
eral years, the growth rate of surveillance videos will be much higher than the
video compression rate that AVC/H.264 and even its next generation development,
HEVC/H.265 (Sullivan et al.
2012
), can achieve. As shown in Fig.
8.1
, this will lead
to a huge gap between them from 2015 to 2020, consequently presenting an unprece-
dented challenge for high-efficiency and low-complexity surveillance video coding
technology.
