Light field image angular domain super-resolution system and method based on zero sample learning

A technology of light field image and sample learning, applied in the field of images, can solve problems such as inability to adapt to different scenes

Active Publication Date: 2020-12-18
BEIHANG UNIV
View PDF5 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the current limited light field data set and the influence of external supervised learning algorithm...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Light field image angular domain super-resolution system and method based on zero sample learning
  • Light field image angular domain super-resolution system and method based on zero sample learning
  • Light field image angular domain super-resolution system and method based on zero sample learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0040] The specific embodiment of the system of the present invention will be further described below in conjunction with the accompanying drawings:

[0041] exist figure 1 In the overall system structure diagram of the present invention, the system of the present invention is divided into a light field image acquisition module, a data preprocessing module, a network training module, a light field image super-resolution module and a data storage module.

[0042] Such as figure 1 As shown, firstly, the light field image L to be super-resolved with an angular domain resolution of 9*9 is obtained through a light field image acquisition module composed of 9*9 cameras distributed on a regular grid and arranged in parallel to the optical axis of the lens. d . In the data preprocessing module, the super-resolved light field image L d Convert from RGB color space to YCbCr color space, and extract its luminance channel as the angle-domain high-resolution light field image for trai...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a light field image angular domain super-resolution system and method based on zero sample learning. The light field image angular domain super-resolution system is composed ofa light field image acquisition module, a data preprocessing module, a network training module, a light field image super-resolution module and a data storage module. According to the system and themethod disclosed in the invention, a zero-sample learning idea is adopted to solve the problem of light field image angular domain super-resolution, and a light field image angular domain super-resolution function is realized in a self-adaptive manner according to the characteristics of different light field images.

Description

technical field [0001] The invention belongs to the field of image technology, and in particular relates to a zero-sample learning-based light field image angle domain super-resolution system and method. Background technique [0002] Traditional cameras can only collect the light intensity information of the scene, but lose the direction information of the light, so the three-dimensional structure information of the scene cannot be obtained. Traditional cameras force computers to observe the three-dimensional world only through two-dimensional images, which has brought bottlenecks to many vision applications. Compared with traditional cameras, light field cameras can simultaneously collect two-dimensional spatial domain and two-dimensional angular domain information of the scene, and use the rich angular domain information to obtain depth information and material information of objects, providing a better solution for scene understanding, virtual reality and other applicatio...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06T3/40
CPCG06T3/4053
Inventor 盛浩王思哲崔正龙杨达周建伟
Owner BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap