Key frame extraction method of motion capture data

Abstract

A key frame extraction method of motion capture data comprises the following steps: (1) feature representation; motion components including rotation angles of a trunk and limbs of a human body and a relative distance between joints of the body are used to represent main features of a human motion posture; (2) critical point screening; an extreme point and a dynamic and static conversion point of a motion component curve serve as a critical point, and a critical point reduction algorithm is used to select a key critical point; (3) key degree curve structure; selected key critical points are used to divide a motion component curve, so as to form a motion component key degree curve; (4) weight learning; and (5) key frame extraction; weighting and fitting are used for a motion component key degree curve of a target motion, and the key frame extraction is carried out based on a fitting curve. The invention does not need to manually set the hard-to-determine threshold, can learn different key frame extraction styles from the sample motion, and can extract the key frame according to several limb motions, so that the need for real-time processing of the motion capture data can be met.

Description

technical field [0001] The invention relates to the comprehensive field of computer animation and machine learning, in particular to a method for extracting key frames of motion capture data based on the key degree of human motion posture. Background technique [0002] Motion capture data is the real human body motion data obtained by motion capture equipment, which is mainly used to drive virtual characters to generate human body animation, and is widely used in animation production, video games, and film and television special effects. Keyframes refer to the 3D pose sequences in motion capture data that can represent human motion, and it provides an important basis for the compression, retrieval, preview, and reuse of large-scale motion capture data. Therefore, how to realize the efficient extraction of key frames of motion capture data has become one of the research hotspots of motion capture technology. [0003] Uniform sampling of motion capture data can be regarded as...

AI Technical Summary

Problems solved by technology

For example, the 2001 paper "Key-posture extraction out of human motion data by curve simplification" by Lim et al. regards each frame of motion data as a point on a high-dimensional space curve, and then uses the curve simplification method to extract Some concave and convex points on the curve are used as key frames. The disadvantage of this method is that the threshold is difficult to determine simply

The 2004 paper "Extracting Key Frames from Motion Capture Data" by Shen Junxing et al. first specifies the first frame of the motion data as the key frame, and then sequentially reduces the subsequent frames whose distance from the key frame is less than the set threshold, and the distance greater than the threshold. The frame is used as a new key frame and continues to be cut. The disadvantage of this method is that it does not take into account the correlation between the cut frame and subsequent key frames.

The paper "Action synopsis: pose selection and illustration" by Assa et al. in 2005 used multi-dimensional scaling to map high-dimensional motion data to low-dimensional space, and then used curve simplification method to extract key frames in low-dimensional space, but this method needs to calculate four Similarity matrix of motion components and data dimensionality reduction, this process is very time-consuming

The paper "Optimization based key frame extraction for motion capture animation" by Liu et al. in 2012 defines the fitness function to measure the reconstruction error between the reconstructed motion and the original motion, with the goal of minimizing the reconstruction error and optimizing the compression rate, and using the genetic algorithm for motion The key frame extraction of data, the shortcoming of this method is that the calculation time is too long, and it cannot meet the needs of real-time processing of motion data

[0004] Although the existing adaptive motion capture data key frame extraction methods effectively overcome the problems in uniform sampling, there are still the following deficiencies in general: (1) Most of the existing methods require the user to manually set various parameters that are difficult to determine. threshold, which causes inconvenience to users

(2) Most of the existing methods extract key frames according to the compression rate and reconstruction error of motion data, but cannot learn different key frame extraction styles based on the key frame samples provided by users

(3) When it is necessary to extract key frames according to the movement of some limbs, most of the existing methods are difficult to meet the needs

Images