Granted Patent B2

US 12,413,769 · App. 18/544,411 · Granted Sep 9, 2025

Inventors: Taoran Lu (Santa Clara, CA); Fangjun Pu (Sunnyvale, CA); Peng Yin (Ithaca, NY); Sean Thomas McCarthy (San Francisco, CA); Tao Chen (Palo Alto, CA)

Assignee: Dolby Laboratories Licensing Corporation

H04N19/513H04N19/105H04N19/172H04N19/33

Abstract

Claims (25)

1. An apparatus to decode a coded bitstream, the apparatus comprising:

an input to receive a coded bitstream; and

a processor to process encoded pictures in the coded bitstream, wherein, for a current picture in the coded bitstream, the processor:

receives a current picture width and a current picture height comprising unsigned integer values;

receives first offset parameters determining a rectangular area on the current picture, wherein the first offset parameters comprise signed integer values;

computes a current area width and a current area height for the rectangular area on the current picture based on the current picture width, the current picture height, and the first offset parameters;

for a reference area, accessing a reference area width, a reference area height, a reference area left offset, and a reference area top offset, wherein accessing the reference area width and the reference area height, further comprises, for a reference picture:

accessing a reference picture width and a reference picture height;

receiving second offset parameters determining a rectangular area in the reference picture, wherein the second offset parameters comprise signed integer values;

computing the reference area width and the reference area height for the rectangular area in the reference picture based on the reference picture width, the reference picture height and the second offset parameters; and

computing the reference area left offset and the reference area top offset based on the second offset parameters;

computes a horizontal scaling factor based on the current area width and the reference area width;

computes a vertical scaling factor based on the current area height and the reference area height;

computes a left-offset adjustment and a top-offset adjustment of the current area based on the first offset parameters; and

performs motion compensation based on the horizontal and vertical scaling factors, the left-offset adjustment, the top-offset adjustment, the reference area left offset, and the reference area top offset.

2. The apparatus of claim 1 , wherein the first offset parameters comprise a left offset, a top offset, a right offset, and a bottom offset.

3. The apparatus of claim 2 , wherein one or more of the left offset, the top offset, the right offset, or the bottom offset comprise values between −2 14 and 2 14 .

4. The apparatus of claim 2 , wherein computing the current area width comprises subtracting from the current picture width a first sum of the left offset and the right offset, and computing the current area height comprises subtracting from the current picture height a second sum of the top offset and the bottom offset.

5. The apparatus of claim 1 , wherein computing the horizontal scaling factor (hori_scale_fp) comprises computing hori_scale_fp=((fRefWidth <<14)+ (fCurWidth>>1))/fCurWidth wherein fRefWidth denotes the reference area width and fCurWidth denotes the current area width.

6. The apparatus of claim 1 , wherein performing motion compensation comprises computing

refxSb L =(((xSb-fCurLeftOffset)<<4)+refMvLX [0]) * hori_scale_fp

refx L =((Sign (refxSb L ) *((Abs (refxSb L )+128) >>8) +XL * ((hori_scale_fp+8) >>4))+32) >>6+ (fRefLeftOffset <<4)

refySbL=(((ySb-fCurTopOffset)<<4)+refMvLX [1]) *vert_scale_fp

refyL=((Sign (refySbL) *((Abs (refySbL)+128) >>8)+yL ((vert_scale_fp+8) >>4))+32) >>6+ (fRefTopOffset <<4)

wherein hori_scale_fp denotes the horizontal scaling factor, vert_scale_fp denotes the vertical scaling fcator, fCurLeftOffset denotes the left-offset adjustment, fCurTopOffset denotes the top-offset adjustment, fRefLeftOffset denotes the reference area left offset, fRefTopOffset denotes the reference area top offset, and (refxSb L , refySb L ) and (refx L , refy L ) are luma locations pointed to by a motion vector (refMvLX [0], refMvLX [1]) given in 1/16-sample units.

Assignments (1)

Continuity (5)

References Cited (62)