A multi-strategy video frame rate improvement method based on spatio-temporal saliency perception

A video frame rate, multi-strategy technology, applied in the field of video processing, can solve the problems of restoration accuracy, computational complexity and human eye attention mismatch, and achieve the effects of good visual quality, low running time, and low computational complexity

Active Publication Date: 2021-09-28
XINYANG NORMAL UNIVERSITY
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Technical problem: Aiming at the problem of the mismatch between the restoration accuracy and computational complexity of the existing technology and the attention of the human eye, a multi-strategy video frame rate improvement method based on spatio-temporal saliency perception is proposed. Obtain high visual satisfaction up-conversion video sequence while calculating the amount

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 multi-strategy video frame rate improvement method based on spatio-temporal saliency perception
  • A multi-strategy video frame rate improvement method based on spatio-temporal saliency perception
  • A multi-strategy video frame rate improvement method based on spatio-temporal saliency perception

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] Further description will be made below in conjunction with the accompanying drawings and the technical solution of the present invention.

[0039] Such as figure 1 Shown, the implementation steps of the present invention are as follows:

[0040] a. Suppose the spatial resolution of the input video sequence is M×N, the total number of frames is L, and the video frame f of the tth frame is t The saliency map for s t . calculate s t The value at the pixel position (i, j) is as follows:

[0041]

[0042] where f t (i,j) is f t Luminance value at pixel position (i, j);

[0043] b. Calculate the salient curve of the video sequence as the value of c at frame t as follows:

[0044]

[0045] normalized, get

[0046]

[0047] Where min{c} is the minimum value in c, and max{c} is the maximum value in c;

[0048] c. Utilize existing technology to construct frame replication, single hypothesis and multi-hypothesis motion estimation modules as follows:

[0049] v ...

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 belongs to the technical field of video processing, and in particular relates to a method for increasing video frame rate by using spatio-temporal saliency detection to adaptively select different motion estimation strategies. The specific steps of the invention are as follows: a. Perform spatio-temporal saliency detection based on brightness comparison on an input video sequence , to obtain the saliency map of each video frame; b. Use the saliency map to quantify the saliency value of each video frame to generate a saliency curve of the video sequence; c. Construct three candidate motion estimation strategies with different restoration accuracy, which are frame replication, single hypothesis Motion estimation and multi-hypothesis motion estimation; d. Calculate the salience value of each interpolation frame, implement an appropriate motion estimation strategy by setting a threshold, and obtain the motion vector field of each interpolation frame; e. Use the motion of each interpolation frame The vector field is interpolated by the overlapping block motion compensation method. The beneficial effect of the present invention is that, according to the salient curve of the video sequence, the motion estimation strategy whose restoration accuracy is adapted to the salient level of each video frame is selected, and the allocation of computing resources is optimized, so that The interpolation quality matches the visual characteristics of human eyes, which improves the satisfaction of human eyes watching videos.

Description

technical field [0001] The invention belongs to the technical field of video processing, and is a conversion method for increasing the video frame rate, in particular to a video frame rate increasing method for adaptively selecting different motion estimation strategies by utilizing spatiotemporal saliency detection. Background technique [0002] With the development of digital high-definition television (HDTV) and high-end multimedia information systems, existing video signal program sources can no longer meet people's needs for visual effects, so it is necessary to increase the frame rate of existing video program sources to achieve better Visual effects; at the same time, as bandwidth resources are continuously allocated and occupied, many video applications can only obtain very limited bandwidth resources (webcasting). In order to meet the requirements of such low-bandwidth applications, it is often necessary to use The frame dropping method reduces the time resolution t...

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): H04N19/182H04N19/186H04N19/577H04N19/587H04N19/59
CPCH04N19/182H04N19/186H04N19/577H04N19/587H04N19/59
Inventor 李然吉秉彧邬长安
Owner XINYANG NORMAL UNIVERSITY
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