Gesture detection method and device, equipment and storage medium

Abstract

The invention discloses a gesture detection method, a gesture detection device, equipment and a storage medium. The gesture detection method comprises the steps of: acquiring an original picture; inputting the original picture into a gesture detection model to obtain prediction annotation information of the original picture, wherein the prediction annotation information of the original picture comprises position information and category probability of a predicted gesture bounding box of the original picture, and the gesture detection model is obtained by balancing weights of a target positivesample and a target negative sample in a training picture in a loss function of a convolutional neural network in the training process of the convolutional neural network; and determining a target gesture bounding box (namely, determining a gesture) from the predicted gesture bounding box of the original picture based on a non-maximum suppression method according to the prediction annotation information of the original picture. According to the gesture detection method and the gesture detection device of the invention, the gesture detection model obtained by balancing the weights of the targetpositive sample and the target negative sample in the training picture in the convolutional neural network is adopted for gesture detection, so that the prediction precision of the gesture detectionmodel on the target gesture detection box is improved.

Description

technical field [0001] Embodiments of the present invention relate to computer vision technology, and in particular to a gesture detection method, device, equipment and storage medium. Background technique [0002] In recent years, with the improvement of computer hardware performance and the emergence of large-scale image data, deep learning has been widely used in the field of computer vision. Among them, the convolutional neural network is a deep learning neural network with outstanding achievements in the field of computer vision. structure. [0003] Gesture detection is a vertical application of object detection in computer vision and is widely used in fields such as human-computer interaction and virtual reality. For gesture detection, gesture detection using a gesture detection model based on convolutional neural network training is widely used. The processing flow of gesture detection based on the gesture detection model generated by convolutional neural network tr...

AI Technical Summary

Problems solved by technology

In the process of training the convolutional neural network to obtain the gesture detection model, due to the insufficient number of positive samples, a large number of bounding boxes that do not contain gestures will be generated. Here, the bounding boxes that do not contain gestures are called negative samples, and a large number of negative samples will be generated. , therefore, there is a category imbalance problem caused by the imbalance in the number of positive and negative samples. The above-mentioned category imbalance problem will make the convolutional neural network unable to be effectively trained, thereby reducing the quality of the gesture detection model generated based on convolutional neural network training. The prediction accuracy of

Images