Game Development Reference
In-Depth Information
Chapter 1
Introduction
1.1 Introduction
We are stepping into the era of big data due to the Internet wave. Besides other
efforts, it is a key challenge to save the cost of storage and transmission of big data.
It is clear that data compression would be a major tool to solve this problem. Within
the large amount of data, multimedia data, including image and video, has become
the majority of Internet data flooding. According to the Cisco VNI report, mobile
video traffic exceeded 50% for the first time in 2011, and Internet video traffic is
predicted to be 86% of all consumer Internet traffic in 2016 (Cisco
2014
). The video
contents created and consumed by people inmodern life are increasing exponentially.
However, video and audio compression efficiency only increases linearly from one
generation to the next. It poses a great challenge to the development of video and
audio compression technologies.
Why video data must be compressed? What is the upper bound of video data
compression? How to compress the video data? What is the constraint of the video
coding standards? What is the state of the art on video coding standards? How to
evaluate the performance of video coding standards? These are the questions we try
to answer in this topic.
The core process of multimedia coding tools, including image coding, video
coding, and audio coding, is to remove the redundancy from multimedia data. The
major part of redundancy inmultimedia data, is fromoversampling during the process
of digitization in spatial space or temporal space. For example, once we set up a
consistent sampling rate at 1,000 samples per second, supposing the signal to be
processed is a sinusoidal wave in 10 cycles per second, then the sampling is 50
times oversampling, according to Nyquist-Shannon sampling theorem (Jerri
1977
).
It is clear that the best way to avoid the oversampling problem is to use an adaptive
sampling strategy according to the frequency of the input signal. But it is unrealistic,
since the signal of image and video data we normally process is fixed from many
kinds of optical signals, it is not possible to use the adaptive sampling method yet,
so consistent sampling is the only selection in today's technology.
