Home >

news Help

Publication Information


Title
Japanese: 
English:A Fine-to-Coarse Convolutional Neural Network for 3D Human Action Recognition 
Author
Japanese: LE THAO MINH, 井上 中順, 篠田 浩一.  
English: Thao Minh Le, Nakamasa Inoue, Koichi Shinoda.  
Language English 
Journal/Book name
Japanese: 
English:Proc. British Machine Vision Conference (BMVC) 
Volume, Number, Page        
Published date Sept. 3, 2018 
Publisher
Japanese: 
English:British Machine Vision Association (BMVA) 
Conference name
Japanese: 
English:29TH BRITISH MACHINE VISION CONFERENCE 
Conference site
Japanese:ニューカッスル 
English:Newcastle 
Official URL http://bmvc2018.org/contents/papers/0745.pdf
 
Abstract This paper presents a new framework for human action recognition from a 3D skeleton sequence. Previous studies do not fully utilize the temporal relationships between video segments in a human action. Some studies successfully used very deep Convolutional Neural Network (CNN) models but often suffer from the data insufficiency problem. In this study, we first segment a skeleton sequence into distinct temporal segments in order to exploit the correlations between them. The temporal and spatial features of a skeleton sequence are then extracted simultaneously by utilizing a fine-to-coarse (F2C) CNN architecture optimized for human skeleton sequences. We evaluate our proposed method on NTU RGB+D and SBU Kinect Interaction dataset. It achieves 79.6% and 84.6% of accuracies on NTU RGB+D with cross-object and cross-view protocol, respectively, which are almost identical with the state-of-the-art performance. In addition, our method significantly improves the accuracy of the actions in two-person interactions.

©2007 Tokyo Institute of Technology All rights reserved.