A Depth Estimation Method for Light Field Image

A technology for depth estimation and light field images, which is applied in the field of depth estimation for light field images, can solve the problems of inaccurate background depth estimation, inability to estimate the depth of main objects and background, abrupt transition of depth change areas, etc., and achieve effective depth information , good stereoscopic perception, enhanced clarity and depth effects

Active Publication Date: 2021-06-04
TIANJIN UNIV
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The existing light field image depth estimation method cannot balance the depth estimation of the main object and the background. It needs to balance the sharpening of the object boundary depth estimation, the abrupt transition of the depth change area inside the object, and the inaccurate background depth estimation.

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
  • A Depth Estimation Method for Light Field Image
  • A Depth Estimation Method for Light Field Image
  • A Depth Estimation Method for Light Field Image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] In order to overcome the deficiencies in the prior art, the embodiment of the present invention uses the depth estimation result obtained by dense stereo matching as a basis, and combines light field structure information to iteratively optimize the global depth in the Markov random field. In the optimization process, the data item considers the depth estimation result based on stereo matching, and the smoothing item combines the gradient information based on the light field structure depth estimation. At the same time, the occlusion area predictor is combined with the energy function smoothing item to effectively reduce the influence of the error depth value of the occlusion area on the global depth optimization process. The specific technical solution mainly includes the following steps:

[0028] 101: Using depth estimation guided by light field structure characteristics, calculate the depth map of the color map of the central viewpoint of the light field image conside...

Embodiment 2

[0037] The scheme in embodiment 1 is further introduced below in conjunction with specific examples and calculation formulas, see the following description for details:

[0038] 1. Depth estimation guided by light field structure characteristics

[0039] During the shooting process of the light field camera, the pixels at corresponding positions in each viewpoint image formed by a certain point in the space scene through the microlens array can be combined into a matrix block, which is called an angle block. Considering that there are holes in the pixels in the light field image, that is, there is no image content consistency in the corresponding angle block, the angle block is divided into the occluded area and the occluded area, and the direction of the dividing line between the two areas and the airspace The boundary direction of the middle occluder is the same. At the same time, since at least half of the pixels in the angle block come from the occluded area, the angle bl...

Embodiment 3

[0102] Combine below figure 2 The scheme in embodiment 1 and 2 is carried out feasibility verification, see the following description for details:

[0103] figure 2 Experimental results of depth estimation from light field images are given. The first row is the depth estimation result of the light field image, and the second row is the left boundary of the smallest hexagonal object in the light field image. From figure 2 It can be seen from the figure that this method can effectively correct the inaccurate depth estimation results in the background, improve the estimation quality of the depth map, and further enhance the clarity and depth perception of the obtained stereoscopic image.

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 method for estimating the depth of a light field image, which includes: using the depth estimation guided by the structural characteristics of the light field to calculate the depth map of the color map of the central viewpoint of the light field image considering the occlusion problem; using the depth map Gradient information is used as the energy function smoothing item to optimize the global depth in the Markov random field framework; multi-scale and multi-window stereo matching is used to calculate the disparity between other viewpoints and the central viewpoint at the same horizontal position as the central viewpoint image; through The method of optical flow interpolates the parallax, and finally the depth estimation result obtained by using the characteristics of the light field image is used as the energy function data item for optimizing the global depth in the Markov random domain framework; considering the presence of occlusion, the occlusion area will affect The object boundary depth estimation combines the occlusion region predictor with the energy function smoothing term, which effectively weakens the influence of the error depth value of the occlusion region on the global depth optimization process.

Description

technical field [0001] The present invention relates to the technical field of image processing and stereo vision, in particular to a depth estimation method for light field images. Background technique [0002] In recent years, light field imaging technology, as a true three-dimensional display technology, has achieved rapid development. Due to the special imaging principle of light field imaging, that is, the light field imaging device can simultaneously record the light position and angle signals in space, making it possible to adjust the depth of field, change the viewpoint, and focus after shooting first. The light field camera is composed of a main lens, a lens array and an imaging sensor, and can obtain the three-dimensional information of the shooting scene through one exposure. At the same time, the light field camera also has the ability to restore the depth information of the shooting scene, which makes the structure of the stereo vision imaging system more conci...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/557
CPCG06T2207/10024G06T2207/10028G06T2207/10052G06T2207/20228G06T7/557
Inventor 雷建军范晓婷侯春萍张凝罗晓维韩梦芯
Owner TIANJIN 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