ENCODING METHOD AND DECODING METHOD OF MOVING IMAGE
PROBLEM TO BE SOLVED: To provide a method of preventing accumulation of rounding errors generated by coprimary interpolation performed in motion compensation, in an image encoding method where the intensity value of brilliance or color difference is quantized to an integer.SOLUTION: With regard to P...
Gespeichert in:
Hauptverfasser: | , |
---|---|
Format: | Patent |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | PROBLEM TO BE SOLVED: To provide a method of preventing accumulation of rounding errors generated by coprimary interpolation performed in motion compensation, in an image encoding method where the intensity value of brilliance or color difference is quantized to an integer.SOLUTION: With regard to P frame performing one-way motion compensation, two types of frames, i.e., P+ frames 503, 507 where a value obtained by adding 0.5 to an integer is rounded up, and P- frames 505, 509 where a value obtained by adding 0.5 to an integer is rounded down, are prepared. A step for decoding a decoded image by using a prediction image is included. When inter-frame prediction is one-way prediction, an image obtained by averaging a first reference image synthesized by motion compensation using plus rounding, and a second reference image synthesized by motion compensation using minus rounding is not used as the reference image of the decoded image. |
---|