Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Directional FMM image restoration method for virtual viewpoint images

A technology of virtual viewpoint and repair method, applied in the field of images with holes, can solve problems such as the influence of the hole area, and achieve the effect of maintaining integrity

Inactive Publication Date: 2015-12-09
杭州益晓步教育科技有限公司
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As a result, the hole area is affected by the pixels in the foreground area

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
  • Directional FMM image restoration method for virtual viewpoint images
  • Directional FMM image restoration method for virtual viewpoint images
  • Directional FMM image restoration method for virtual viewpoint images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] The present invention will be further described below in conjunction with accompanying drawing

[0018] A large hole will be generated around the foreground of the new virtual viewpoint image after DIBR. These holes should be filled with the pixel values ​​of the background area as much as possible. Therefore, there will be two situations: for the drawn left virtual viewpoint image, the foreground area is shifted to the right, and the hole is generated on the left side of the foreground area. At this time, the pixel value of the background area on the left side of the hole is sampled to fill the hole area. The reverse is true for the right virtual viewpoint image.

[0019] Such as figure 2 For the right virtual viewpoint image drawn, the hollow area of ​​the icon shadow should be filled with the background pixels to the right of the hollow. In this way, when there are no multiple reference viewpoints or other conditions are not met, only the image inpainting algorit...

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 directional FMM image restoration method for virtual viewpoint images. For a new viewpoint image obtained after depth image based rendering, image restoration is carried out by a directional fast marching method to obtain a new viewpoint image with clear foreground edge. A hole is repaired from left to right for a left virtual viewpoint image, and a hole is repaired from right to left for a right virtual viewpoint image. According to the method of the invention, first, two types of expansion nucleuses are used to perform similar T-shaped expansion and rectangular expansion of a hole marking mask figure, and an original mask figure is subtracted to obtain a narrow edge and a wide edge of a single side; then, the distance from each pixel in the broad edge of the single side to the hole edge is solved; and finally, image restoration is carried out in a particular direction from the narrow edge. Therefore, an effect different from that of a traditional fast marching method is achieved, namely, a large hole can be repaired well. Compared with the traditional fast marching method, the foreground edge information can be kept better by the method of the invention.

Description

technical field [0001] The invention belongs to the field of hole image repair, and relates to a directional FMM image repair method for a virtual viewpoint image, in particular to an image with holes obtained after drawing a virtual viewpoint based on a depth map. Background technique [0002] With the development of 3D technology and the gradual enhancement of 3D hardware equipment, people's demand for 3D products is increasing day by day. This enables the development of 2D to 3D technology, and also makes the model-based rendering technology (ModelBasedRendering, MBR) and image-based rendering technology (ImageBasedRendering, IBR) become the key technology of virtual view synthesis. Image-based rendering technology has gradually become mainstream due to its fast rendering speed and small bandwidth requirements. Among them, DepthImageBasedRendering (DIBR) is widely used. For the new viewpoint image drawn by DIBR technology, there will be a series of visibility problems, ...

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): H04N13/00
Inventor 杨柏林王聪聪
Owner 杭州益晓步教育科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com