Human action recognition in RGB-D videos using motion sequence information and deep learning
Human action recognition in RGB-D videos using motion sequence information and deep learning
复制标题
DOI：
10.1016/j.patcog.2017.07.013
复制
发表时间：
2017-12-01
影响因子：
8
通讯作者：
Chalavadi, Krishna Mohan
中科院分区：
文献类型：
--
作者：
Ijjina, Earnest Paul;Chalavadi, Krishna Mohan
在本文中，我们提出了一种使用深度学习基于 RGB-D 视频中的运动序列信息来识别人类动作的方法。提出了一种新的表示方法，强调与每个动作相关的关键姿势。从 RGB 和深度视频流中的运动获得的特征作为卷积神经网络的输入来学习判别特征。该方法的有效性在 MIVIA 动作、NATOPS 手势、SBU Kinect 交互和 Weizmann 数据集上得到了证明。 (C) 2017 Elsevier Ltd. 保留所有权利。
In this paper, we propose an approach for recognizing human actions based on motion sequence information in RGB-D video using deep learning. A new representation that gives emphasis to the key poses associated with each action is presented. The features obtained from motion in RGB and depth video streams are given as input to the convolutional neural network to learn the discriminative features. The efficacy of the proposed approach is demonstrated on MIVIA action, NATOPS gesture, SBU Kinect interaction, and Weizmann datasets. (C) 2017 Elsevier Ltd. All rights reserved.