Human Action Recognition Based on Improved Two-Stream Convolution Network
Open Access
- 7 June 2022
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 12 (12) , 5784
Abstract
Two-stream convolution network (2SCN) is a classical method of action recognition. It is capable of extracting action information from two dimensions: spatial and temporal streams. However, the method of extracting motion features from a spatial stream is single-frame recognition, and there is still room for improvement in the perception ability of appearance coherence features. The classical two-stream convolution network structure is modified in this paper by utilizing the strong mining capabilities of the bidirectional gated recurrent unit (BiGRU) to allow the neural network to extract the appearance coherence features of actions. In addition, this paper introduces an attention mechanism (SimAM) based on neuroscience theory, which improves the accuracy and stability of neural networks. Experiments show that the method proposed in this paper (BS-2SCN, BiGRU-SimAM Two-stream convolution network) has high accuracy. The accuracy is improved by 2.6% on the UCF101 data set and 11.7% on the HMDB51 data set.Keywords
Funding Information
- nuist students’ platform for innovation and entrepreneurship training program (202110300120Y)
This publication has 65 references indexed in Scilit:
- Large-Scale Video Classification with Convolutional Neural Networks2014
- Effective 3D action recognition using EigenJointsJournal of Visual Communication and Image Representation, 2014
- Action Recognition with Improved Trajectories2013
- Dense Trajectories and Motion Boundary Descriptors for Action RecognitionInternational Journal of Computer Vision, 2013
- Recognizing 50 human action categories of web videosMachine Vision and Applications, 2012
- Visual attention: The past 25 yearsVision Research, 2011
- Free viewpoint action recognition using motion history volumesComputer Vision and Image Understanding, 2006
- Early and Late Mechanisms of Surround Suppression in Striate Cortex of MacaqueThe Journal of Neuroscience, 2005
- Long Short-Term MemoryNeural Computation, 1997
- Distributed representations, simple recurrent networks, and grammatical structureMachine Learning, 1991