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

A 2d to 3d depth estimation method

A depth estimation and 3D technology, which is applied in the field of image processing and stereo vision, can solve problems such as depth estimation errors, and achieve accurate and high-quality parallax estimation

Active Publication Date: 2019-06-04
TIANJIN UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the semi-automatic depth estimation method can obtain a high-quality key frame depth map, the non-key frame depth map obtained by the depth propagation algorithm has errors in the depth estimation of moving objects, and it is necessary to perform depth analysis on the non-key frame depth map Optimized for higher quality depth video sequences

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 2d to 3d depth estimation method
  • A 2d to 3d depth estimation method
  • A 2d to 3d depth estimation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0047] 1. Keyframe Parallax Assignment

[0048] The basic processing unit in the process of semi-automatic 2D to 3D conversion is a video sequence, that is, a series of image frames with continuous time and relatively little change in spatial objects. Usually the first and last frames of the video are marked as key frames, and the rest of the image frames are regarded as non-key frames. Manually mark the depth of keyframes, assign different disparity values ​​to different objects in the input color keyframe image, and generate marked disparity curves, which are regarded as strokes marking clues as sparse disparity maps.

[0049] 2. Key frame disparity estimation based on non-local random walk

[0050] The marked curve strokes and the key frame color map are used as the input of the NRW algorithm, and the NRW algorithm is used to segment the image to obtain an image composed of objects with different disparity values, which is used as the disparity map of the key frame. The N...

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 present invention belongs to the image processing and stereoscopic vision technology field, and relates to a method by employing manual labelling for performing semi-automatic depth estimation of 2D video to obtain high-quality depth video sequence so as to generate a 3D video and generate a 3D video based on the 2D video. According to the invention, the method comprises the following steps: 1, key frame parallax distribution; 2, key frame parallax estimation based on non-local random walk; 3, parallax propagation base on the mobile bilateral filtering; and 4, non-key frame depth optimization based on the non-local random walk. The 2D-to-3D depth estimation method is mainly applied to the image processing and stereoscopic vision application occasions.

Description

technical field [0001] The invention belongs to the technical field of image processing and stereo vision, and relates to a method for performing depth estimation on the problem of semi-automatic 2D to 3D conversion of a monocular view. Background technique [0002] Due to the high requirements for shooting accuracy in the collection process of depth information, the shooting is difficult and takes a long time, which makes the production process of 3D video more complicated, which leads to the limitation of the growth of 3D content and affects the development of 3D industry. The 2D to 3D technology takes into account the existing rich 2D video / image resources, uses the texture structure information in it for depth estimation, and obtains the corresponding depth map. Among them, depth estimation, that is, obtaining depth information in a scene from a 2D video / image is a key part of the 2D to 3D technology. [0003] According to the acquisition method of depth information, th...

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): G06T7/55
CPCG06T2207/10016G06T2207/20228
Inventor 雷建军张凝侯春萍张翠翠郑凯夫丛润民
Owner TIANJIN UNIV
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