Hand key point recognition model training method and device, recognition method and device

Abstract

The invention discloses a hand key point recognition model training method and device and a recognition method and device, and belongs to the field of gesture recognition. The method comprises the following steps of: using a Cycle-GAN model to convert the sample virtual image into a simulation image, wherein the sample virtual image is an image generated by three-dimensional modeling, the samplevirtual image comprises key point coordinates corresponding to hand key points, and the simulation image is used for simulating an image collected under a real scene; extracting a hand image in the simulation image; and training a hand key point recognition model according to the hand image and the key point coordinates in the simulation image, the hand key point recognition model being used for outputting the hand key point coordinates of the hand in the real image according to the input real image. In the embodiment of the invention, the training sample is closer to the acquired real image,and the accuracy of gesture recognition on the real image by using the hand key point recognition model subsequently is higher.

Description

technical field [0001] The embodiments of the present application relate to the technical field of gesture recognition, and in particular to a method for training a recognition model of key points of a hand, a recognition method and a device. Background technique [0002] With the continuous maturity of artificial intelligence technology, more and more application scenarios begin to support human-computer interaction, and gesture interaction is a common way of human-computer interaction. [0003] The key to gesture interaction is hand key point recognition. In a hand key point recognition method, the developer first constructs a virtual 3D model of the hand through a computer, and uses the virtual image of the virtual 3D model and the 3D coordinate data corresponding to the key points of the hand as training samples to train the key points of the hand. point recognition model. In the subsequent gesture recognition, after inputting the real image containing gestures collect...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application provides a recognition model training method, recognition method and equipment for key points of the hand, which can solve the problem that the recognition model based on virtual image training in the related art has a low accuracy rate of gesture recognition and affects the accuracy of gesture interaction. sexual problems

Images