A hybrid approach to keyframe extraction from motion capture data using curve simplification and principal component analysis

Abstract

In this paper, we propose a novel method to extract keyframes from motion capture data. A hybrid approach, which combines a curve‐simplification algorithm with an initialization procedure including principal component analysis, is adopted. The developed method automatically extracts an appropriate number of keyframes at high speed without performance degradation. Experimental results prove the effectiveness of the present method. © 2014 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

KeyFrame extraction for human motion capture data via multiple binomial fitting

“…Curve simplification takes a frame as a point and a motion sequence as a curve in a high-dimensional space, and realizes keyframe extraction by finding a series of points that can depict the whole curve well. [5][6][7] However, the keyframe is extracted according to the local extremum of the motion curve, and other parts of the curve with obvious slope change are ignored. The method based on matrix decomposition is to represent motion sequence as motion matrix, and then decompose the matrix into weight matrix and keyframe matrix approximately, 8,9 but this method is time-consuming and easy to ignore the time information, and the extracted keyframes may not cover the whole sequence.…”

“…Linear interpolation and quaternion interpolation are commonly used in motion reconstruction. 5,6,12,14 In animation, linear interpolation and quaternion interpolation are very good for intermediate frame compensation. Using the data fitting method to correct the motion capture error caused by deleting force feedback can effectively reduce the traditional motion error and has good application value.…”

“…The methods of keyframe extraction mainly include curve simplification‐based, matrix decomposition‐based, cluster‐based, and optimization‐based. Curve simplification takes a frame as a point and a motion sequence as a curve in a high‐dimensional space, and realizes keyframe extraction by finding a series of points that can depict the whole curve well 5‐7 . However, the keyframe is extracted according to the local extremum of the motion curve, and other parts of the curve with obvious slope change are ignored.…”

“…Linear interpolation and quaternion interpolation are commonly used in motion reconstruction 5,6,12,14 . In animation, linear interpolation and quaternion interpolation are very good for intermediate frame compensation.…”

Optimal and interactive keyframe selection for motion capture

“…Techniques have been proposed that use extrema of the first principle component of the animation data [10], that pick keypoints using curvature-related finite differences operating at a coarse scale [11], that consider changes between neighbouring poses [12][13][14], and others that use a saliency measure based on differences between Gaussian-weighted averages at different scales [15,16]-this measure can also be interpreted as a derivative, recalling a classic vision result that the difference of Gaussians closely approximates the Laplacian of a Gaussian-smoothed version of the signal [17].…”