Video Content AdaptationQoS by using the tree algorithm

1
Video Content AdaptationQoS by using the tree
algorithm

• 2003. 4. 9.
• 20022161 Jaepil Han
• 20022156 Jinho Choo
• 20022081 Sunwoo Yang

2
Content

• Introduction
• Related Theory
• Content Adaptation
• - Framwork
• - Implementation
• The present works
• Further works

3
Introduction(1/2)

• The optimal tradeoff point between transcoding
  overhead(CPU cost) and storage needed for the
  various pre-processed content variants
• Content adaptation involves creating different
  versions suitable for rendering by mobile devices
  from some original content such as a web page.
• The decision engine will look for the best
  tradeoff among various parameters in order to
  reduce the loss of quality in various domains.

4
Introduction(2/2)

• The decision engine will take into account the
  different types of context such as device
  capability profile, networking charateristics,
  user preference, and metadata about the content
  itself
• Well implement the engine for bandwidth among
  networking charateristics.
• Study the video content adaptation and
  implementation it by using the Tree algorithm

5
Related thoery(1/2)

• Trees algorithm
• - is a connected undirected graph with no
  simple circuits
• - particulary useful in computer science
• for instance, it can be used to construct
  efficient codes for storing and transmitting data

6
Related thoery(2/2)

• - this algorithm is applied to the decision
  engine for the best tadeoff among various
  parameters in order to reduce the loss of quality
  in various domains.
• for example, frame dropping, coefficient
  dropping, PSNR, Bandwidth.

The Applied Tree algorithm
7
The content Adaptation (1/5)

• Framework

8
The content Adaptation (2/5)

• Frame dropping (FD)
• Selection of the skipped frame locations
• Based on frame type or perceptual importance of
  the frames
• Consider decoding-independent cases only
• The first B, all B, all BP, P frames dropping
• Only a coarse level of target bit rate is
  achievable

9
The content Adaptation (3/5)

• DCT coefficient dropping (CD)
• Meet the target bit rate with a greater
  granularity
• Percentage of rate-reduction to be achieved like
  a CD (10)
• Minimize the overall distortion by optimally
  allocating the number of dropped coefficients
  among blocks
• find breakpoints that
  minimize the CD distortion while meeting the
  target rate as
• where and denote the rate
  required to encode block and the distortion,
  respectively, when the breakpoint value is
  used, and is the total number of blocks

10
The content Adaptation(4/5)

• Implementation
• - Procedure

11
The content Adaptation(5/5)

• Reference Info
• - reference value according to Bandwidth
• FD(Frame Dropping), CD(Coefficient
  Dropping), PSNR
• Parsing
• - reference Info is xml file format. So this
  file must be parsed
• Decision Making
• - find adapted parameter by tree algorithm,
  Interpolation, Logic.
• Adapt the bit stream parameter
• - generate the adpated bit stream parameter
  file

12
The present works(1/2)

• Implement Parsing system

(a) Top xml file (b) Bottom parsed data
13
The present works(2/2)

• Implement the Interface
• - input the User Info(bandwidth)
• - display the value of the adapted parameter

14
Futher works

• In Adaptation engine,
• - Implementation of Decision making
• Find efficiently and accurately Optimized
  
• value
• - adapt the bit stream