Gesture recognition method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides a gesture recognition method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps: obtaining an original image which comprises a gesture; performing hand target detection on the original image to obtain a hand key feature data set; training a preset initial gesture recognition model by using the hand key feature data set as a training data set to obtain a basic gesture recognition model; pruning the basic gesture recognition model to obtain a standard gesture recognition model; and obtaining a to-be-recognized image, and recognizing a gesture type in the to-be-recognized image by using the standard hand recognition model. According to the gesture recognition method provided by the invention, the gesture recognition model is combined, the features of each gesture type are deeply mined, the trained gesture recognition model is pruned and optimized according to task requirements, and the accuracy of the model is improved on the premise of not additionally increasing computing resources.

Description

technical field [0001] The present application relates to the technical fields of computer vision and deep learning, and in particular, to a gesture recognition method, apparatus, electronic device, and computer-readable storage medium. Background technique [0002] Gesture recognition is one of the most popular human-computer interaction methods, which can be applied to smart home, smart transportation, virtual reality, smart TV, online education, aerial keyboard, somatosensory games, smart cockpit, smart fitness mirror, smart logistics, etc. in various fields. [0003] Due to the diversity of gestures themselves, there are many types to be recognized, and in real-time dynamic scenarios, gestures will also be accompanied by motion blur during the switching process, so the gesture recognition algorithm will be inaccurate and robust in technical applications. Sex is not strong. In order to improve the recognition accuracy, some researchers have proposed some network structu...

AI Technical Summary

Problems solved by technology

[0003] Due to the diversity of the gesture itself, there are many types that need to be recognized, and in the real-time dynamic scene, the gesture will be accompanied by motion blur during the switching process, so the accuracy of the gesture recognition algorithm in the technical application is not high, and the robustness Sex is not strong

In order to improve the recognition accuracy, some researchers have proposed some network structures with high complexity, such as SPP-Net, R-CNN, FasterRCNN, etc., but due to the large amount of calculation of the model, these are not suitable for deployment on the mobile terminal, and some researchers Use a lightweight network, but the accuracy cannot achieve the effect of the algorithm landing

Images