Human brain visual memory principle-based human body action identification method and system

Abstract

The invention discloses a human brain visual memory principle-based human body action identification method and system, and relates to the field of computer vision and video monitoring. Enlightened by the human brain visual memory principle, the invention proposes the following technical scheme for the first time: in a training stage, feature coding of local features is used to train a classifier model, and is used to build a visual memory bank; in an identification stage, the feature coding of the local features of a video to be identified is searched in the visual memory bank; and part of the local features of the video in a search result are used to replace shielded information in the video to be identified, feature coding is performed on local features of the replaced video, and the training model is input to perform testing, thereby obtaining the category of a human body action in the video. The human brain visual memory principle-based human body action identification method and the system can effectively solve the problem of shielding in human body action identification.

Description

technical field [0001] The present invention relates to the fields of computer vision and video surveillance, in particular to a method and system for human action recognition based on the principle of human brain visual memory. Background technique [0002] Video-based human action recognition is a very important problem with applications in video surveillance, video retrieval, and human-computer interaction. Human action recognition refers to the use of computers to distinguish the categories of human actions from video sequences. [0003] Video-based human action recognition can be divided into two parts: action representation and action classification. Videos can be divided into training set and testing set. Action representation refers to extracting appropriate feature data from a video sequence containing human action to describe the action of the human body. Classification of actions refers to obtaining a classifier model by learning the feature data in the trainin...

AI Technical Summary

Problems solved by technology

Probabilistic-based methods, such as Bayesian networks, hidden Markov models, are flat models and are effective in representing simple actions, but cannot describe the hierarchical and shared structure in complex actions

Pose-based methods need to use detectors to train each body part by manually annotating training images, which limits the application of pose-based methods in action recognition.

Images