METHOD AND APPARATUS FOR ENCODING VIDEO USING GRANULAR DOWNSAMPLING OF FRAME RESOLUTION

Patent №

US 8,780,976

Granted

2014-07-15

Filed 2011

Owner

GOOGLE INC.

AI components

1

vision

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

13096285

Disclosed herein are methods and apparatuses for encoding a video signal having a plurality of frames, each frame having a plurality of blocks and an original resolution. The method includes identifying a first frame from the plurality of frames that is predicted using inter prediction, determining a first variance for the first frame using a processor. selecting a first frame resolution from one of: (a) the original resolution or (b) a resolution lower than the original resolution for the first frame based on the first variance, and encoding the first frame using the selected first frame resolution.

VisionH04N 19/176H04N 19/132H04N 19/15H04N 19/503

AI classification

Vision1.00
Knowledge representation0.34
AI hardware0.08
Natural language0.00
Evolutionary computation0.00
Planning0.00
Machine learning0.00
Speech0.00

Ownership

GOOGLE INC.

assignment · 261950535

Assignors

JIA, WEI

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC