SWITCHING DISTORTION METRICS DURING MOTION ESTIMATION

Patent №

US 8,155,195

Granted

2012-04-10

Filed 2006

Owner

MICROSOFT CORPORATION

AI components

1

vision

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

11400049

Techniques and tools for switching distortion metrics during motion estimation are described. For example, a video encoder determines a distortion metric selection criterion for motion estimation. The criterion can be based on initial results of the motion estimation. To evaluate the criterion, the encoder can compare the criterion to a threshold that depends on a current quantization parameter. The encoder selects between multiple available distortion metrics, which can include a sample-domain distortion metric (e.g., SAD) and a transform-domain distortion metric (e.g., SAHD). The encoder uses the selected distortion metric in the motion estimation. Selectively switching between SAD and SAHD provides rate-distortion performance superior to using only SAD or only SAHD. Moreover, due to the lower complexity of SAD, the computational complexity of motion estimation with SAD-SAHD switching is typically less than motion estimation that always uses SAHD.

AI classification

Vision1.00
AI hardware0.01
Evolutionary computation0.00
Natural language0.00
Speech0.00
Machine learning0.00
Knowledge representation0.00
Planning0.00

Ownership

MICROSOFT CORPORATION

assignment · 178670001

Assignors

REGUNATHAN, SHANKAR, LIN, CHIH-LUNG, HOLCOMB, THOMAS W.

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

© 2026 NYSGPT2525 LLC