Human Action Recognition Based on Dense Trajectories Analysis and Random Forest  

作　　者：Pin-Zhong Pan Chung-Lin Huang 

机构地区：[1]Sonix Technology,HsinChu 30265,HsinChu [2]Department of M-Commerce and Multimedia Applications,Asia University,Taichung 41354,Taichung

出　　处：《Journal of Electronic Science and Technology》2016年第4期370-376,共7页电子科技学刊（英文版）

基　　金：supported by the MOST,Taiwan under Grant No.102-2221-E-468-013

摘　　要：This paper presents a human action recognition method. It analyzes the spatio-temporal grids along the dense trajectories and generates the histogram of oriented gradients （HOG） and histogram of optical flow （HOF） to describe the appearance and motion of the human object. Then, HOG combined with HOF is converted to bag-of-words （BoWs） by the vocabulary tree. Finally, it applies random forest to recognize the type of human action. In the experiments, KTH database and URADL database are tested for the performance evaluation. Comparing with the other approaches, we show that our approach has a better performance for the action videos with high inter-class and low inter-class variabilities.This paper presents a human action recognition method. It analyzes the spatio-temporal grids along the dense trajectories and generates the histogram of oriented gradients （HOG） and histogram of optical flow （HOF） to describe the appearance and motion of the human object. Then, HOG combined with HOF is converted to bag-of-words （BoWs） by the vocabulary tree. Finally, it applies random forest to recognize the type of human action. In the experiments, KTH database and URADL database are tested for the performance evaluation. Comparing with the other approaches, we show that our approach has a better performance for the action videos with high inter-class and low inter-class variabilities.

关 键 词：Bag-of-words （BoWs） dense trajectories histogram of optical flow （HOF） histogram of oriented gradient （HOG） random forest vocabulary tree. 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]