Typo in 8.5.5.7

[abe.kiyo]

In 8.5.5.7 Derivation process for luma affine control point motion vector predictors, the neighboring sample locations are defined as follows:

3.The sample locations ( xNbA0, yNbA0 ), ( xNbA1, yNbA1 ), ( xNbA2, yNbA2 ), ( xNbB0, yNbB0 ), ( xNbB1, yNbB1 ), and ( xNbB2, yNbB2 ) are derived as follows:

( xA0, yA0 ) = ( xCb − 1, yCb + cbHeight ) (811)
( xA1, yA1 ) = ( xCb − 1, yCb + cbHeight − 1 ) (812)
( xB0, yB0 ) = ( xCb + cbWidth , yCb − 1 ) (813)
( xB1, yB1 ) = ( xCb + cbWidth − 1, yCb − 1 ) (814)
( xB2, yB2 ) = ( xCb − 1, yCb − 1 ) (815)

I would suggest to fix it as follows:

3.The sample locations ( xNbA0, yNbA0 ), ( xNbA1, yNbA1 ), ( xNbA2, yNbA2 ), ( xNbB0, yNbB0 ), ( xNbB1, yNbB1 ), and ( xNbB2, yNbB2 ) are derived as follows:

( xNbA0, yNbA0 ) = ( xCb − 1, yCb + cbHeight ) (811)
( xNbA1, yNbA1 ) = ( xCb − 1, yCb + cbHeight − 1 ) (812)
( xNbB0, yNbB0 ) = ( xCb + cbWidth , yCb − 1 ) (813)
( xNbB1, yNbB1 ) = ( xCb + cbWidth − 1, yCb − 1 ) (814)
( xNbB2, yNbB2 ) = ( xCb − 1, yCb − 1 ) (815)

[jlchen]

Good catch! This issue will be fixed in S2001-vH release.