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

SLAM method of RGB-D camera in dynamic scene

A dynamic scene, RGB image technology, applied in the field of computer vision, can solve the problems of time-consuming, high computational cost, and difficult to run.

Active Publication Date: 2020-11-10
SOUTH CHINA UNIV OF TECH +1
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The method based on deep learning has quite good performance in highly dynamic scenes. However, this method uses a target detection algorithm based on deep learning and a pixel-level object segmentation algorithm, which leads to high computational overhead, is very time-consuming, and is also difficult to real-time run on CPU

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
  • SLAM method of RGB-D camera in dynamic scene
  • SLAM method of RGB-D camera in dynamic scene
  • SLAM method of RGB-D camera in dynamic scene

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0099] Such as figure 1 As shown, the present embodiment specifically discloses a SLAM method of an RGB-D camera in a dynamic scene, the method comprising the following steps:

[0100] S1, ORB feature points (such as figure 2 As shown) extract and calculate the descriptor of the extracted ORB feature points, match the RGB image of the current frame with the RGB image of the previous frame, and initially calculate the pose of the current frame by minimizing the reprojection error;

[0101] S2. Compare the pose of the current frame with the poses of all key frames, try to find out the first m key frames whose poses are similar to the current frame and far from the current frame in time, and set them as reference key frames set, expressed as:

[0102]

[0103] Among them, a, b, c, d are constants greater than 0, id is the serial number of the key frame, id c is the serial number of the current frame, R is the rotation matrix of the key frame, t is the translation matrix of...

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 an SLAM method of an RGBD camera in a dynamic scene, so that the RGBD camera effectively overcomes the influence of a dynamic object, and the positioning precision of the RGBDcamera in the dynamic scene is improved. In order to detect feature points on a dynamic target in real time, the invention provides a dynamic feature point detection mechanism of double clustering. Then, the likelihood that the feature points are static is estimated simultaneously in geometric and temporal dimensions. In the geometric dimension, a static probability is created for each feature point of the current frame to indicate that the feature point is static; in the time dimension, whether the feature points are static feature points or not is judged according to all frames rather than the current frame. Finally, in order to enable the RANSAC algorithm to be more suitable for a dynamic environment, the traditional RANSAC algorithm is improved. Experimental results show that the positioning precision of the RGBD camera in a dynamic environment is effectively improved, and the RGBD camera can be operated on a CPU in real time.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a SLAM method of an RGB-D camera in a dynamic scene. Background technique [0002] After years of development, SLAM (simultaneous localization and map construction) technology has been very mature and applied to many fields. Traditionally, most visual SLAM systems are assumed to operate in a static environment. However, in the real application scenarios of SLAM, there are many dynamic objects, such as on the factory transportation line, there are many workers walking back and forth; In the mall there are always people walking around and so on. Therefore, most of the scenes in real life are dynamic, which also means that many SLAM systems are difficult to work in the actual dynamic environment. [0003] Therefore, many researchers have done a lot of research work on SLAM in dynamic environments. Generally, there are three methods that can be used to reduce the error of c...

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): G06K9/32G06K9/62
CPCG06V20/62G06F18/2321G06F18/2415Y02T10/40
Inventor 刘屿潘文钊邬依林
Owner SOUTH CHINA UNIV OF TECH
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