Unlock instant, AI-driven research and patent intelligence for your innovation.

A traffic video deshaking method based on feature trajectory

A video deshaking and trajectory technology, applied in the field of computer vision, can solve problems such as algorithm failure and time-consuming 3D method

Active Publication Date: 2021-03-09
UNIV OF SCI & TECH OF CHINA
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the 3D method is time-consuming, and it is easy to cause algorithm failure when the parallax is not obvious

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 traffic video deshaking method based on feature trajectory
  • A traffic video deshaking method based on feature trajectory
  • A traffic video deshaking method based on feature trajectory

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0079] Such as figure 1 Shown, the inventive method comprises the steps:

[0080] The step of feature track generation is to extract feature points from the traffic video frame by frame, and obtain the corresponding feature track according to feature point matching;

[0081] The optimization function of time and space smoothes the camera path (ie, the feature trajectory), and the optimization function based on time and space smoothes the feature trajectory obtained in Step1 to obtain a smooth feature trajectory; use distributed optimization in the smoothing process method;

[0082] The step of affine transformation from a shaken view to a stable view is to perform an affine transformation from a shaken view to a stable view frame by frame based on the feature trajectory before and after smoothing.

[0083] The specific implementation of the above steps will be described in detail below.

[0084] 1. Feature trajectory generation

[0085] 1.1 Continuous Trajectory Extraction ...

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 relates to a traffic video deshaking algorithm based on foreground and background feature trajectories. Traffic video is more challenging than general handheld video recording, including higher frequency video jitter, more foreground objects, greater occlusion, and more serious parallax. The present invention regards the problem of deshaking traffic video as the smoothing problem of camera path. The camera motion is estimated using all the foreground and background feature trajectories, and the camera path is smoothed by an optimization function based on time and space. The method proposed by the present invention does not need to distinguish between foreground and background trajectories, so the error caused by this step will not be introduced. The more comprehensive estimation also effectively reduces the error caused by only using background information for deshaking and affine estimation, as well as the distortion at the foreground and disparity. The improved performance of the algorithm is especially noticeable when dealing with traffic videos with large foreground and severe parallax.

Description

technical field [0001] The invention relates to a traffic video de-shaking algorithm based on foreground and background feature trajectories, computer vision, and video de-shaking. Background technique [0002] In recent years, more and more cameras have been used in various scenes in real life, including a large number of portable and movable camera equipment, hand-held equipment and vehicle-mounted equipment due to artificial shaking or engine shaking caused the recorded video The sharp drop in quality and violent shaking caused people's sensory discomfort. Traffic video has some particularities compared with general handheld device recording video: first, severe vehicle shaking causes traffic video to have higher frequency jitter; second, traffic video has more foreground motion; third, The problem of occlusion of large foreground objects is easy to occur in traffic videos. [0003] At present, there are relatively few studies on traffic video de-shaking. [1,2] used som...

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): G06T5/00
CPCG06T5/002G06T2207/10016G06T2207/20021
Inventor 凌强赵敏达
Owner UNIV OF SCI & TECH OF CHINA