Group behavior identification method based on channel information fusion and group relationship spatial structural modeling

Abstract

The invention provides a group behavior identification method based on channel information fusion and group relationship spatial structural modeling. The method comprises the following steps: firstly,segmenting a to-be-identified video, sampling a plurality of frames at equal intervals, and extracting fusion features containing space-time and motion information through an improved STM network module; performing intra-frame region division and high-dimensional mapping on the fusion feature of each frame to form graph structure data; and finally, through a graph convolution-LSTM network containing a core group relationship evolution model, integrating a global behavior discrimination feature and a local behavior discrimination feature as a group behavior descriptor to discriminate behaviorclassification, and obtaining a final behavior label through softmax. According to the scheme, a channel selection module is added to fuse space and motion features, so that feature representation containing space and motion information is extracted at the same time, and the relevance of the features is enhanced; the spatial structural modeling of the group relationship is combined to ensure the integrity and comprehensiveness of the extracted spatiotemporal information features, and the key object of the group interaction relationship which plays a decisive role in behavior discrimination isemphatically considered, so that the recognition precision can be effectively improved.

Description

technical field [0001] The invention relates to the field of group behavior recognition, in particular to a group behavior recognition method based on channel information fusion and group relationship space structured modeling. Background technique [0002] In recent years, human action recognition in videos has achieved remarkable achievements in the field of computer vision. Behavior analysis has also been widely used in real life, such as intelligent video surveillance, abnormal event detection, sports analysis, understanding social behavior, etc. These applications make behavior recognition have important scientific practicability and huge economic value. As deep learning has gradually achieved great success in the field of computer vision, neural networks have gradually been applied to video-based human behavior recognition, and have achieved remarkable results. [0003] For example, the scheme named "Region based multi-stream convolutional neural networks for collecti...

AI Technical Summary

Problems solved by technology

In the feature extraction part of this scheme, although not only the local spatial information and motion information of the key figures are extracted, but also the global spatial information and motion information of the entire image are extracted, the scheme is only fused at the last layer of the network. The method cannot aggregate spatial and motion information very well, and only using a dual-stream network can only capture short-term information, while group behavior recognition in video emphasizes long-term time-series information, so the behavior recognition accuracy of this scheme is not high. ideal

Images