Patent · US Active

Face hallucination using convolutional neural networks

US9405960B2 · kind B2 · utility

9Cited by

2References

18Claims

0Family size

Assignee

BEIJING KUANGSHI TECHNOLOGY CO., LTD. · CN

Inventors

Qi Yin · Beijing, CN
Zhimin Cao · Beijing, CN
Erjin Zhou · Beijing, CN

Key dates

Filing date	Jun 17, 2014
Grant date	Aug 2, 2016
Priority date	—
Expiry date	Jul 14, 2034

Classification

Technology area (CPC G)Physics
CPC primaryG06V40/171
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Face hallucination using a bi-channel deep convolutional neural network (BCNN), which can adaptively fuse two channels of information. In one example, the BCNN is implemented to extract high level features from an input image. The extracted high level features are combined with low level details in the input image to produce the higher resolution image. Preferably, a proper coefficient is obtained to adaptively combine the high level features and the low level details.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.