Video Action Detection Method Based on Scale Attention Dilated Convolutional Network

Abstract

The invention discloses a video action detection method based on a scale attention hole convolution network. The method of the present invention first samples the video to obtain the frame image sequence and obtains the video segment according to the segment position mark, then constructs the layer-scale attention action segment model and the frame position attention action recognition model respectively, and obtains the frames sequentially according to the above model and in combination with the watershed algorithm The weighted feature representation of the image and the action category to which the video clip belongs to complete the task of video action detection. The method of the present invention utilizes the dilated convolutional network to extract the spatio-temporal motion information that can better reflect the intrinsic structure of the time dimension and the space dimension of the video data, and more properly characterizes the internal correlation of the timing context of the video frame with the change of the scale size through layer-scale attention , the designed frame position attention mechanism gives the video frame of the action segment a weight that more accurately characterizes the key content of the action segment, improves the accuracy of video action detection, and improves the efficiency of action detection.

Description

technical field [0001] The invention belongs to the technical field of video analysis, in particular to the technical field of time series action detection, and relates to a video action detection method based on a scale-attention hole convolution network. Background technique [0002] The understanding of human motion video plays an important role in many fields such as security monitoring and behavior analysis, and has become a frontier research topic in the field of computer vision. However, unedited real videos often contain background segments unrelated to human actions, which will affect the correct understanding of video content. To solve this problem, the video action detection method not only classifies the actions in the video, but also locates the start and end time of the action instance in the video. Video action detection tasks usually use video frame sequences as input, and output the detection results of multiple groups of segments in the form of "action typ...

AI Technical Summary

Problems solved by technology

[0004] The existing video action detection methods mainly have the following deficiencies: First, in the feature extraction stage, the three-dimensional convolution operation used to extract the timing features of the action will fixedly reduce the timing dimension of the input video layer by layer in the constructed network model, Constrains the scale size of the extracted features in time series. Too small a scale may cause contextual semantic segmentation, and too large a scale may cause interference of different semantics. The key points of whether or not and its type, that is, the key frame position and its duration (such as continuous key frames) are often different, and the conventional average pooling operation ignores the weight of the key points; third, the existing methods for different scales Fragments use different network structures (such as dilated convolutional networks) to extract feature representations of action clips, which will greatly increase the time and space costs of network construction and training

Images