Human Action Recognition Based on Improved Two-Stream Convolution Network

• Published in: Applied sciences Vol. 12; no. 12; p. 5784
• Main Authors: Wang, Zhongwen, Lu, Haozhu, Jin, Junlan, Hu, Kai
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.06.2022
• Subjects: Accuracy; action recognition; Algorithms; BiGRU network; Coherence; Convolution; Datasets; Deep learning; Human activity recognition; Human motion; Hypotheses; Memory; Motion perception; Nervous system; Neural networks; Neurosciences; Parameter identification; SimAM attention mechanism; two-stream convolution network
• ISSN: 2076-3417; 2076-3417
• DOI: 10.3390/app12125784

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.