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

A method for automatic scoring of figure skating videos based on deep learning

A deep learning and video technology, applied in the direction of instruments, calculations, character and pattern recognition, etc., can solve the problems of low accuracy, two kinds of score prediction of video features, and insufficient utilization, so as to achieve fast training and reduce the length of the input sequence Effect

Active Publication Date: 2020-12-22
FUDAN UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this method uses the structure of deep neural network, it does not make full use of the information in the video, and the obtained video features are not enough to accurately predict the two scores respectively, so the accuracy is low

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 method for automatic scoring of figure skating videos based on deep learning
  • A method for automatic scoring of figure skating videos based on deep learning
  • A method for automatic scoring of figure skating videos based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Step 1. Collect and label figure skating videos. When collecting videos, you should first ensure that the scoring standards for videos from different competitions are the same. For this reason, we only collect related videos from the past 5 years; Videos of different contestants selected from a series of events such as Japan Station (NHK), China Cup World Figure Skating Grand Prix (CoC) and so on. Each video corresponds to the scoring of nine judges. The resulting 500 videos contained 149 different players from 20 countries. On this basis, we collect the total technical score (TES) and program content score (PCS) corresponding to each video;

[0045] Step 2. Preprocess the collected videos and extract low-order feature sequences. Since it is more complicated to use the entire video as the input of the deep neural network, it is generally input in the form of an image sequence. Therefore, the present invention decodes and extracts frames from the video to obtain a se...

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 a method for automatically scoring figure skating videos based on deep learning. According to the definition of total technology score (TES) and program content score (PCS) of figure skating and the different aspects targeted, the present invention proposes a local information extraction module of self-attention mechanism and a multi-scale convolutional neural network based on the idea of ​​deep learning. The global information extraction module of the network, and combine these two modules to propose an automatic scoring method for figure skating videos based on video segment-level features. This method has the advantages of high precision and good robustness, and is not only suitable for figure skating, but also for other sports that are scored based on technical movements and overall performance.

Description

technical field [0001] The invention belongs to the technical field of computer video analysis, and in particular relates to a method for automatically scoring figure skating videos based on deep learning. Background technique [0002] Video has become an essential part of human life. In many fields including security, robotics, entertainment, etc., video provides convenient services for human beings with the rich information it can convey, and the practicality of video makes recording, viewing and dissemination more and more widespread. Along with video promotion, the manpower and material resources required to watch and analyze a large number of videos has also become a thorny issue. To solve this problem, an effective solution is to use machines instead of humans to automatically extract useful information from videos for analysis. Therefore, video understanding has become a topic that has received more attention in computer vision, and video scoring is a more specific ...

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): G06K9/00G06K9/62
CPCG06V40/20G06V20/42G06V20/46G06F18/214
Inventor 付彦伟徐程明姜育刚薛向阳
Owner FUDAN UNIV
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