Adaptive filtering based upon boundary strength

摘要

Adjacent regions are identified in an image. Coding parameters for the adjacent regions are identified. Selective filtering is performed at the region between the identified adjacent regions.

主权项

1. A method for at least one of encoding and decoding a digital image with an encoding or decoding apparatus, comprising:
identifying adjacent regions in said digital image; determining, for said adjacent regions, whether each region is intra coded; determining whether or not to filter, with a filter of said apparatus, at least a portion of digital image data in said regions proximate a boundary between said adjacent regions based on whether at least one of said adjacent regions is intra coded, where said determining whether or not to filter includes determining whether or not to filter said portion of digital image data based on motion vectors associated with said adjacent regions; establishing a boundary strength of said boundary by
setting said boundary strength to a first value in response to at least one of said adjacent regions being intra coded; andsetting said boundary strength to a a second value, which is less than the first value in response toboth said adjacent regions being not intra coded;neither of said adjacent regions having a non-zero transform coefficient; andeither the absolute value of the difference of horizontal components of motion vectors in said adjacent regions or the absolute value of the difference of vertical components of motion vectors in said adjacent regions being greater than or equal to a predetermined threshold value; and controlling a strength of said filter according to said boundary strength such that, in response to at least one of said adjacent regions being intra coded, the strength of the filter for the boundary is always stronger than the filter for other cases.