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

Prediction method for video saliency

A predictive method, a remarkable technology, applied in image analysis, image enhancement, instrumentation, etc., to solve the effect of adaptive lighting

Active Publication Date: 2020-11-17
STATE GRID ZHEJIANG ELECTRIC POWER +1
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, current models cannot theoretically guarantee that the detected salient regions can best restore the image both locally and globally.

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
  • Prediction method for video saliency
  • Prediction method for video saliency
  • Prediction method for video saliency

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] The present invention provides a prediction method for video salience, characterized in that the prediction method includes:

[0045] Step 1: Construct the initialization saliency map, calculate the salient area in the saliency map for quantitative calculation, and obtain the quantitative sparse matrix response result;

[0046] Step 2, build a decomposition matrix, perform local coordinate constraints on each region in the feature space based on the decomposition matrix, and perform saliency map calculation based on the local coordinate constraints;

[0047] Step 3, calculate the saliency map for each frame in the video, and perform quality prediction.

[0048] In practice, the technical problem to be solved by the present invention is to provide a method for predicting area-level saliency from a video using a camera that records the position of the human line of sight. Using a camera-assisted NMF algorithm can ensure that the detected saliency area can be Describe a 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 provides a prediction method for video saliency, belonging to the image detection field, comprising the steps of constructing an initialization saliency map, calculating a saliency areain the saliency map for quantization calculation, and obtaining a quantization sparse matrix response result; the factorization matrix is constructed, which is used to constrain the local coordinatesof each region in the feature space, and the saliency map is calculated based on the local coordinates constraint conditions. The saliency map is calculated for each frame in the video, and the quality prediction is carried out. A non-negative matrix factorization algorithm is proposed based on the current popular camera applications. The optimized video saliency is calculated by combining the learned human line-of-sight orientation with the initialized saliency map.

Description

technical field [0001] The invention belongs to the field of image detection, in particular to a prediction method for video salience. Background technique [0002] In general, humans can select visual information by moving their gaze, bringing the high resolution of the retina to the most relevant areas of the video. In this way, humans can process relevant visual information and interpret complex scenes in real time with limited resources. At present, with the development of digital technology, video data has shown explosive growth, so fast semantic analysis and information acquisition of video are becoming more and more important. [0003] With the current popularity of embedded cameras (such as iMAC or notebook cameras), the camera-assisted mechanism of the present invention improves saliency prediction by discovering human-engaged regions in video clips. Building a computational model that can effectively predict visually / semantically salient regions in a video is a 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
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/00
CPCG06T7/0002G06T2207/10016G06T2207/30168
Inventor 姚一杨罗旺王彦波叶伟静欧阳柳章振海
Owner STATE GRID ZHEJIANG ELECTRIC POWER