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

Method for extracting key frame of human motion sequence

A human motion and key frame technology, which is applied in the fields of computer graphics and human-computer interaction, can solve the problems of difficulty in optimizing the core parameters of clustering algorithms, difficulty in extracting high-quality key frames, and difficulty in measuring the similarity between frames.

Active Publication Date: 2018-09-11
BEIJING UNIV OF TECH
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the human motion sequence has the space-time dimension and the frequency is variable. Not only is it difficult to measure the similarity between frames, but also the core parameters of the clustering algorithm are also difficult to optimize.
Existing clustering algorithms such as K-means clustering are difficult to predict the number of clusters, and the algorithm is unstable with respect to the initial cluster center, so it is difficult to achieve high-quality key frame extraction

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
  • Method for extracting key frame of human motion sequence
  • Method for extracting key frame of human motion sequence
  • Method for extracting key frame of human motion sequence

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] Such as Figure 8 As shown, the present invention provides a method for extracting key frames of a human motion sequence, comprising the following steps:

[0017] (1) Human motion frame-to-frame similarity measurement based on joint point cluster feature representation

[0018] The human body motion capture data of CMU is composed of 38 joint points, but considering that some nodes have little influence on motion semantic analysis, we choose a simplified skeleton model composed of 18 joint points (such as figure 1 shown). Its 3D coordinates can be expressed as:

[0019] f(t)={θ 1 (t), θ 2 (t),K,θ 18 (t)}, t ∈ {1,2,K,T} (1)

[0020] where θ i (t)=(x i (t),y i (t),z i (t)), i ∈ {1,2,K,18}, θ i (t) represents the 3D coordinate information of the i-th human body joint point in the t-th frame, and T represents the frame number.

[0021] By calculating the offset information of each joint point and other joint points in each frame as the relative position feature ...

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 extracting a key frame of a human motion sequence. For a given human motion sequence, position offset vectors of each joint point and other joint points of each frame are firstly calculated and taken as features of the corresponding joint points; then, according to the prior knowledge of human motions, the joint point group is defined in the form of joint pointcombination, the feature combination of the joint point group realizes the feature representation of the motion frame, and at the same time, a similarity measure model of human motion frames based onthe feature representation of the joint point group; and finally, based on this model, the affine propagation clustering algorithm is used to realize data-adaptive extraction of the key frame of thehuman motion sequence.

Description

technical field [0001] The invention belongs to the technical fields of computer graphics and human-computer interaction, and in particular relates to a method for extracting key frames of human motion sequences. Background technique [0002] With the rapid development of sensor technology and the continuous expansion of the application field of human-computer interaction technology, it has become a feasible and inevitable task to obtain the joint space-time position information of the moving human body; and the high-frequency sampling acquisition in the time domain required by the high complexity of human motion , making the human motion data sequence naturally have high redundancy, which directly leads to difficulties in storage, retrieval, browsing, and reuse of human motion data. Extracting representative frames or key frames from a given human motion sequence to realize the compressed representation of the human motion sequence at the semantic level is the key technolog...

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 Applications(China)
IPC IPC(8): G06K9/00H04N5/14
CPCH04N5/144G06V40/23
Inventor 孔德慧孙彬王少帆王玉萍王立春
Owner BEIJING UNIV OF TECH