A Method based on Spatial and Temporal Multi Grained Information for Gait Recognition

Gait recognition is one of the new biometric techniques. However, gait recognition is easily affected by external factors, resulting in a decrease in the accuracy. Therefore, this paper proposes a gait recognition method based on spatial and temporal multi grained information. This method uses hierarchical network to obtain different receptive fields, focusing on both shallow and deep features of gait. In addition, more spatial and temporal information can be saved to improve the accuracy of gait recognition. Finally, the proposed method is experimentally verified on the CASIA-B dataset. the experimental results show that the accuracy of the proposed method is greatly improved under normal walking, carrying bags and wearing coats.