A recursive scene understanding method based on multiple semantic interactions

A semantic interaction and scene understanding technology, applied in the field of recursive scene understanding based on multiple semantic interactions, can solve problems such as system errors that cannot be corrected, achieve good scalability and improve deficiencies

Inactive Publication Date: 2019-02-15
NINGBO UNIVERSITY OF TECHNOLOGY
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the main defect of this idea is that once a certain link makes an e

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 recursive scene understanding method based on multiple semantic interactions
  • A recursive scene understanding method based on multiple semantic interactions
  • A recursive scene understanding method based on multiple semantic interactions

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] A recursive scene understanding method based on multiple semantic interactions of the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0029] Such as figure 1 As shown, a recursive scene understanding method based on multiple semantic interactions, including steps:

[0030] S1. Provide an image to be detected, and output the geometric and semantic intrinsic images of the image to be detected through scene surface layout estimation;

[0031] S2. According to the reasoning of the physical boundary in the auxiliary scene of the above-mentioned geometric and semantic intrinsic image, and estimate the relative depth relationship of the object in the image scene by combining the inferred object boundary information and camera viewpoint information;

[0032] S3. Perform object / viewpoint detection on the image to be detected, and combine the estimated results of the depth relationship to obtain the fi...

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 provides a recursive scene understanding method based on multiple semantic interactions. The method comprises the following steps: providing an image to be detected; estimating and outputting geometric and semantic eigenimages of the image to be detected through the surface layout of the scene; and outputting the geometric and semantic eigenimages of the image to be detected. According to the geometrical and semantic eigenimages, the relative depth of the object in the scene is estimated by combining the inferred object boundary information and the camera viewpoint information. Object/viewpoint detection is performed on the image to be detected, and the final detection result is obtained by combining the estimation result of depth relationship. This feedback design based on eigeninformation interaction can effectively improve the shortcomings of the feedforward system, and has good scalability.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a recursive scene understanding method based on multiple semantic interactions. Background technique [0002] As a research hotspot in the field of computer vision, scene understanding technology has received extensive attention and research. According to the multi-level semantic representation of images, scene understanding can be divided into two aspects: local scene understanding and global scene understanding. Among them, the former involves a variety of semantic research contained in images such as object class and shape recognition, camera pose and position estimation, and scene depth prediction, while the latter focuses on the overall analysis of the scene, which requires comprehensive semantic mining of various images. Based on the analysis of the inherent correlation between different semantics, that is, the study of the image context relationship. The research on the i...

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): G06K9/00G06K9/62
CPCG06V20/10G06F18/24G06F18/214
Inventor 姚拓中安鹏何加铭
Owner NINGBO UNIVERSITY OF TECHNOLOGY
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