Behavior analysis model training method, behavior analysis method and device

Abstract

The invention discloses a behavior analysis model training method, a behavior analysis method and equipment thereof. The model training method comprises the following steps: constructing a continuous frame human eye training data set; generating a human eye cluster image based on the continuous frame human eye training data set; inputting the human eye cluster image into a to-be-trained analysis model to obtain an eye behavior classification of a continuous frame human eye training data set, the eye behavior classification including an eye behavior category and a first prediction probability thereof; and obtaining a first prediction loss of the to-be-trained analysis model based on the eye behavior classification of the continuous frame human eye training data set and the real eye behavior tag, and training the to-be-trained analysis model by using the first prediction loss to obtain a final behavior analysis model. According to the behavior analysis model training method, the plurality of human eye image frames of the continuous frame human eye training data set are coded to the same human eye cluster image, so that parameter redundancy, complex calculation and storage consumption of model training can be effectively reduced, and the efficiency of model training is improved.

Description

technical field [0001] The present application relates to the technical field of digital image recognition, in particular to a behavior analysis model training method, a behavior analysis method and equipment thereof. Background technique [0002] Blinking behavior is a normal physiological response that naturally reflects biological activity. In order to avoid eye fatigue and reflect the good interaction between people and the outside world, it is inevitable to blink. In real life, with the popularization of smart technology, blinking behavior has been widely used in scenarios such as liveness detection, fatigue monitoring, micro-expression analysis, and polygraph systems. [0003] However, the current academic achievements of eye blink detection based on single-frame spatial features lack timing information, and there is a natural problem of information loss for behavior recognition; the academic achievements of eye blink detection based on sensing equipment and devices h...

AI Technical Summary

Problems solved by technology

[0003] However, the current academic achievements of eye blink detection based on single-frame spatial features lack timing information, and there is a natural problem of information loss for behavior recognition; the academic achievements of eye blink detection based on sensing equipment and devices have high equipment costs and difficult application scenarios. Extended to general scenes, the problem of low practicability; the academic achievements of blink detection based on multi-frame timing weighted fusion have the problem of complex calculation and storage consumption, which will make the algorithm more difficult to use

Images