Patent №
US 8,311,318
Granted
2012-11-13
Filed 2010
Owner
CHUNGHWA PICTURE TUBES, LTD.
Lab
—
AI components
1
vision
Assignment
Recorded
Dataset
AIPD
2023_r1 edition
Application
12952591
The present invention provides a system for generating images of multi-views. The system includes a processing unit; an image range calculating module coupled to the processing unit to calculate the ranges of a background image and a main body image of a 2D original image of an article; a depth model generating module coupled to the processing unit to generate a depth model according to an equation; an image cutting module coupled to the processing unit to cut the 2D original image of the article or the depth model to generate a cut 2D image of the article or a depth model with a main body image outline; a pixel shifting module coupled to the processing unit to shift every pixel in the main body image of the 2D original image of the article according to the depth model with the main body image outline to obtain shifted main body images of multi-views; and an image synthesizing module coupled to the processing unit to synthesize the shifted main body images of multi-views and background figures of multi-views to obtain final images of multi-views for 3D image reconstruction.
AI classification
Ownership
CHUNGHWA PICTURE TUBES, LTD.
assignment · 253970651
Assignors
KAO, MENG-CHAO, CHEN, CHIEN-HUNG, SHEN, TZU-CHIANG
On an employer assignment, the assignors are typically the inventors.