Patent №
US 7,899,115
Granted
2011-03-01
Filed 2008
Owner
LG ELECTRONICS INC.
Lab
—
AI components
1
vision
Assignment
Recorded
Dataset
AIPD
2023_r1 edition
Application
11918441
The present invention relates to a method for scalably encoding and decoding a video signal. In an embodiment of the present invention, motion information, such as reference indices, motion vector, modes, etc., for an enhanced layer can be derived from a base layer, and a reference index for an image block in the enhanced layer is determined using median criterion based upon the base layer. A reference index for a block, having the most pixels corresponding to those of the image block, in the base layer is selected as the reference index for the image block, and the nearer reference index is selected if more than one block in the base layer have the same pixels corresponding to those of the image block. And, a motion vector for the base layer block related to the selected reference index is selected as the motion vector for the image block.
AI classification
Ownership
LG ELECTRONICS INC.
assignment · 216940314
Assignors
JEON, BYEONG MOON, PARK, SEUNG WOOK, PARK, JI HO, YOON, DOE HYUN
On an employer assignment, the assignors are typically the inventors.