Dance Video Movement Extraction Through Skeleton Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional dance video processing methods require significant human resources for manually selecting and creating standard movement sequences, which is inefficient and affects the conversion efficiency of user-uploaded dance videos into game content.
Innovation Solution
A method and apparatus that utilizes a skeleton node recognition algorithm to extract and cluster dance movements from user-uploaded videos, identifying key skeleton node images to generate standard movement sequences through cluster analysis, ensuring accurate and efficient conversion into game-compatible formats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual selection and creation of standard movement sequences is used, then accuracy and precision of movement extraction is improved, but productivity and conversion efficiency deteriorate
Solution Approach 1:
The system performs automatic skeleton node recognition and cluster analysis on user-uploaded dance videos to extract standard movement sequences without requiring manual annotation. The algorithm autonomously processes the entire workflow from video input to movement sequence output, enabling the system to serve itself rather than requiring human intervention for each step.
Solution Approach 2:
The patent replaces the manual mechanical process of selecting and creating movement sequences with an automated computational system. Skeleton node recognition algorithms automatically detect dancer positions and poses, while cluster analysis algorithms automatically group similar movements, substituting human expertise with computational methods that scale efficiently.
2Productivity
If automated skeleton node recognition is used, then productivity and conversion efficiency is improved, but device complexity and processing requirements worsen
Solution Approach 1:
The patent segments the complex video processing task into distinct modular components: skeleton node recognition module, cluster analysis module, and standard sequence generation module. Each module handles a specific sub-task independently, making the overall complex system more manageable and easier to implement through standardized processing stages.
Solution Approach 2:
The patent introduces skeleton node images as an intermediary representation between the raw video frames and the final standard movement sequences. This intermediate format simplifies the processing by converting complex video data into standardized skeletal representations that can be easily clustered and processed, acting as a mediator that reduces processing complexity.
3Measurement precision
If cluster analysis is performed on all skeleton node images, then measurement precision and accuracy of movement classification is improved, but loss of time and processing duration worsen
Solution Approach 1:
The patent applies cluster analysis to a selective subset of skeleton node images rather than processing every single frame. By identifying and analyzing only the most representative and distinctive movement patterns, the system achieves high classification accuracy while reducing processing time, avoiding the excessive action of analyzing all images in detail.
Data Source
AI summary
This disclosure provides techniques of extracting movements from a dance video. The techniques comprise receiving a dance video that includes one or more dancers and that is uploaded by a user, and obtaining a video frame in the dance video; recognizing, based on a skeleton node recognition algorithm, skeleton node images from the video frames corresponding to a target dancer, where the target dancer is selected from the one or more dancers; performing cluster analysis on the skeleton node images recognized from the dance video and corresponding to each target dancer to obtain a plurality of cluster sets; determining a cluster center in each of the plurality of cluster sets as a key skeleton node image; and sequentially outputting the key skeleton node images to obtain a standard movement sequence corresponding to each target dancer in the dance video.