BriefGPT.xyz
Apr, 2016
基于核的协方差方法进行动作识别
Kernelized Covariance for Action Recognition
HTML
PDF
Jacopo Cavazza, Andrea Zunino, Marco San Biagio, Vittorio Murino
TL;DR
本文提出了一种基于核技巧的协方差矩阵增强方法Kernelized-COV，可用于建模复杂的非线性关系，通过在3D动作识别基准数据集上的实验证明了其优越性。
Abstract
In this paper we aim at increasing the descriptive power of the
covariance matrix
, limited in capturing linear mutual dependencies between variables only. We present a rigorous and principled mathematical pipeline to recover the
→