Video encoding device and video encoding method
AssigneeMitsubishi Electric Corporation
InventorAtsumasa Osawa
Life of this patent
Estimating the term…
Abstract
Disclosed is a video encoding device including an arithmetic contents determining unit that takes into consideration the arithmetic contents of an SAO arithmetic operation performed on a local decoded image of an already-encoded block pointed to by a motion vector which is searched for through a motion-compensated prediction process performed by a motion-compensated prediction unit, to determine the arithmetic contents of an SAO arithmetic operation to be performed on a local decoded image after a deblocking filter process performed by a deblocking filter unit.
Claims6
A video encoding device comprising:
- a frame memory to hold a local decoded image of an already-encoded block;
- a prediction image generator to perform a motion-compensated prediction process on a coding target block by using the local decoded image held by said frame memory, to generate a prediction image of said coding target block;
- an encoder to compress a difference image between said coding target block and said prediction image, and encode compressed data about said difference image;
- a local decoded image generator to decompress said compressed data and decode said difference image, and generate a local decoded image from the decoded difference image and said prediction image;
- an arithmetic contents determinator to take into consideration arithmetic contents of an SAO (sample adaptive offset) arithmetic operation performed on a local decoded image of an already-encoded block pointed to by a motion vector which is searched for through the motion-compensated prediction process performed by said prediction image generator, to determine arithmetic contents of an SAO arithmetic operation to be performed on the local decoded image generated by said local decoded image generator after deblocking filtering;
- and
- an SAO arithmetic operator to perform an SAO arithmetic operation according to the arithmetic contents determined by said arithmetic contents determinator on the local decoded image generated by said local decoded image generator, and store the local decoded image after the SAO arithmetic operation in said frame memory.
A video encoding method comprising the steps of:
- in a prediction image generator, performing a motion-compensated prediction process on a coding target block by using a local decoded image of an already-encoded block, the local decoded image being held by a frame memory, to generate a prediction image of said coding target block;
- in an encoder, compressing a difference image between said coding target block and said prediction image, and encoding compressed data about said difference image;
- in a local decoded image generator, decompressing said compressed data and decoding said difference image, and generating a local decoded image from the decoded difference image and said prediction image;
- in an arithmetic contents determinator, taking into consideration arithmetic contents of an SAO arithmetic operation performed on a local decoded image of an already-encoded block pointed to by a motion vector which is searched for through said motion-compensated prediction process, to determine arithmetic contents of an SAO arithmetic operation to be performed on the local decoded image generated in said local decoded image generation processing after deblock filtering;
- and
- in an SAO arithmetic operator, performing an SAO arithmetic operation according to the arithmetic contents determined in said arithmetic contents determination processing step on the local decoded image generated in said local decoded image generation processing step, and storing the local decoded image after the SAO arithmetic operation in said frame memory.
Dependencies are inferred from the claim text.