US Patent Application 18143156. VIRTUAL BOUNDARY PROCESSING FOR ADAPTIVE LOOP FILTERING simplified abstract

From WikiPatents
Jump to navigation Jump to search

VIRTUAL BOUNDARY PROCESSING FOR ADAPTIVE LOOP FILTERING

Organization Name

Telefonaktiebolaget LM Ericsson (publ)


Inventor(s)

Kenneth Andersson of Gävle (SE)

[[Jacob Str�m of Stockholm (SE)]]

Zhi Zhang of Solna (SE)

Jack Enhorn of Kista (SE)

VIRTUAL BOUNDARY PROCESSING FOR ADAPTIVE LOOP FILTERING - A simplified explanation of the abstract

This abstract first appeared for US patent application 18143156 titled 'VIRTUAL BOUNDARY PROCESSING FOR ADAPTIVE LOOP FILTERING

Simplified Explanation

The abstract describes a method for encoding or decoding an image using a coding tree block (CTB) and selecting a filter strength value for each sample in the CTB based on its location within the block.

  • The method involves obtaining a CTB, which is a block of sample values associated with the image.
  • Each sample value in the CTB has a location (x,y) within the block, where x is the horizontal value and y is the vertical value.
  • The method includes selecting a filter strength value for each sample based on its vertical location (y) within the CTB.
  • The filter strength value is determined by checking if a specific condition is true: (y==(CtbSizeY−5) OR y==(CtbSizeY−4)) AND var==1.
  • If the condition is true, a filter strength value of 10 is chosen for the sample.
  • The method allows for filtering of the samples in the CTB based on their location and the selected filter strength value.


Original Abstract Submitted

A method for encoding or decoding an image. The method includes obtaining a coding tree block (CTB) having a vertical size (CtbSizeY) and a horizontal size (CtbSizex), the CTB comprising set of sample values associated with the image, the set of sample values comprising a current sample value having a location (x,y) within the CTB, where x is a horizontal value and y is a vertical value. The method also includes selecting a filter strength value for the current sample based on y and filtering the current sample using the selected filter strength value. Selecting a filter strength value for the current sample based on y comprises: i) determining whether the following first condition is true: (y==(CtbSizeY−5) OR y==(CtbSizeY−4)) AND var==1, where var is a variable that is set equal to 1 as a result of a determination that a second condition is satisfied; and ii) choosing a filter strength value of 10 as a result of determining that the first condition is true.