Invention Grant
- Patent Title: Image upsampling using global and local constraints
-
Application No.: US15059629Application Date: 2016-03-03
-
Publication No.: US09836820B2Publication Date: 2017-12-05
- Inventor: Oncel Tuzel , Yuichi Taguchi
- Applicant: Mitsubishi Electric Research Laboratories, Inc.
- Applicant Address: US MA Cambridge
- Assignee: Mitsubishi Electric Research Laboratories, Inc.
- Current Assignee: Mitsubishi Electric Research Laboratories, Inc.
- Current Assignee Address: US MA Cambridge
- Agent Gene Vinokur; James McAleenan; Hironori Tsukamoto
- Main IPC: G06T3/40
- IPC: G06T3/40 ; G06K9/66 ; G06T5/00 ; G06K9/62 ; G06K9/52 ; H04N19/136 ; G06K9/00

Abstract:
A method upsamples an image using a non-linear fully connected neural network to produce only global details of an upsampled image and interpolates the image to produce a smooth upsampled image. The method concatenates the global details and the smooth upsampled image into a tensor and applies a sequence of nonlinear convolutions to the tensor using a convolutional neural network to produce the upsampled image.
Public/Granted literature
- US20170256033A1 Image Upsampling using Global and Local Constraints Public/Granted day:2017-09-07
Information query