Human body gesture letter recognition method and device, computer device and storage medium

Abstract

According to the human body gesture letter recognition method and device, a computer device and a storage medium, character input is conducted by adopting a brand-new recognition mode of combining motion information and image information according to sign language habits of a sign language user, the standard sign language action standard is met, operation is easy, and use is convenient.

Description

technical field [0001] The present invention relates to the field of image recognition technology and wearable technology, in particular to a hand-worn gesture recognition device and method applied to recognize human gesture letters, and the method can further recognize sign language words by expanding the sample feature label library and other meaningful gestures. Background technique [0002] Smart wearable devices are a hot topic in the current engineering world, especially those that provide human-computer interaction are even more sought after by people. From the launch of Google Glass to the release of the first smart watch, wearable devices with various functions have entered people's lives so far, but the current methods of text input, especially Chinese input, in wearable devices But very limited. Now there are mainly three input methods: voice input, touch screen and other on-screen keyboard or physical keyboard input, and relatively popular gesture input. [00...

AI Technical Summary

Problems solved by technology

[0003] Compared with the above three input methods, voice input has certain restrictions on the user group and the environment in which it is used. Deaf-mute people who cannot speak and can only use sign language and other language dysfunction groups are greatly hindered in using it, and when there is noise The accuracy rate is not high in the environment, and the degree of environmental constraints is relatively large, and the sound transmission will have a certain range, which cannot protect the user's privacy very well.

Traditional physical keyboards or on-screen keyboards are restricted by physical volume. Physical keyboards are large and inconvenient to carry, while on-screen keyboards are limited by screen volume. If the screen is small, the correct click rate will be greatly reduced.

The current body recognition technology is mostly used in large-scale gesture motion capture and recognition, and is mostly used in large-scale body motion capture such as body fall alarm and somatosensory game virtual entertainment. At present, the only product based on text input is Ring. The principle is to place a sensor on one finger, and use this finger to input English with continuous strokes. Because English handwriting has fewer strokes, and Chinese strokes are more complicated, it is actually difficult to implement Chinese input based on continuous handwriting.

[0004] At present, there is no suitable wearable smart device for Chinese text input.

Images