Difference between revisions of "H.264 Prediction"

From MultimediaWiki
Jump to navigation Jump to search
(→‎Diagonal Down/Left: SVQ3 doesn't use this precise mode)
(add a placeholder for the special SVQ3 4x4 mode)
Line 69: Line 69:
   f = (T5 + 2*T6 + T7 + 2) / 4
   f = (T5 + 2*T6 + T7 + 2) / 4
   g = (T6 * 3*T7      + 2) / 4
   g = (T6 * 3*T7      + 2) / 4
=== Diagonal Down/Left (SVQ3) ===


=== Diagonal Down/Right ===
=== Diagonal Down/Right ===

Revision as of 16:49, 29 July 2007

This page documents the various prediction methods used in H.264 and related formats such as Sorenson Video 3 and RealVideo 4.

4x4 Prediction Modes

Vertical

  • H.264: mode 0
  • SVQ3: mode 0
  • RV40: mode 0
  D | T0  T1  T2  T3
---------------------
 L0 | T0  T1  T2  T3
 L1 | T0  T1  T2  T3
 L2 | T0  T1  T2  T3
 L3 | T0  T1  T2  T3

Horizontal

  • H.264: mode 1
  • SVQ3: mode 1
  • RV40: mode 1
  D | T0  T1  T2  T3
---------------------
 L0 | L0  L0  L0  L0
 L1 | L1  L1  L1  L1
 L2 | L2  L2  L2  L2
 L3 | L3  L3  L3  L3

DC

  • H.264: mode 2
  • SVQ3: mode 2
  • RV40: mode 2
  D | T0  T1  T2  T3
---------------------
 L0 |  Z   Z   Z   Z
 L1 |  Z   Z   Z   Z
 L2 |  Z   Z   Z   Z
 L3 |  Z   Z   Z   Z


where Z is computed as:

Z = (T0 + T1 + T2 + T3 + L0 + L1 + L2 + L3 + 4) / 8

Diagonal Down/Left

  • H.264: mode 3
  • SVQ3: not used
  • RV40: mode 3
  D | T0  T1  T2  T3  T4  T5  T6  T7
-------------------------------------
 L0 |  a   b   c   d
 L1 |  b   c   d   e
 L2 |  c   d   e   f
 L3 |  d   e   f   g

where:

 a = (T0 + 2*T1 + T2 + 2) / 4
 b = (T1 + 2*T2 + T3 + 2) / 4
 c = (T2 + 2*T3 + T4 + 2) / 4
 d = (T3 + 2*T4 + T5 + 2) / 4
 e = (T4 + 2*T5 + T6 + 2) / 4
 f = (T5 + 2*T6 + T7 + 2) / 4
 g = (T6 * 3*T7      + 2) / 4

Diagonal Down/Left (SVQ3)

Diagonal Down/Right

Vertical/Right

Horizontal/Down

Vertical/Left

Horizontal/Up

Left/DC

Top/DC

DC-128

16x16 Prediction Modes

DC

Vertical

Horizontal

Plane

Left Mean