Gesture control method and system

Abstract

The embodiment of the invention provides a gesture control method. The method comprises the following steps: in response to a wake-up word input by a user, starting a camera to acquire a gesture of the user; and judging whether to output a corresponding action or not according to the acquired dynamic gesture of the user. The embodiment of the invention further provides a gesture control system forthe electronic equipment. According to the embodiment of the invention, gesture recognition is started after voice wake-up instead of starting camera shooting when the television is turned on, so that CPU use of the television in a non-wake-up state at ordinary times is effectively reduced. And meanwhile, considering that the static gesture is easy to mistakenly trigger, a more accurate dynamic gesture is used, false triggering and false recognition are avoided, the gesture hit rate is improved, and the user experience is improved. Considering that different scenes may have different gestures, the scenes are divided, and the dynamic gestures are further limited through the scenes, so that different recognition results and states exist in different scene modes, and multi-mode control is realized.

Description

technical field [0001] The invention relates to the field of intelligent voice, in particular to a gesture control method and system. Background technique [0002] With the development of intelligent voice technology, more and more smart devices have the function of voice control. In order to improve the efficiency of interaction with users, gesture control is introduced on the basis of voice interaction to improve the interaction experience. [0003] For example, the voice combined with the gesture of the static picture realizes the control of the TV set, such as: after the voice wakes up the TV set, the static gesture ok realizes the playback operation of a certain selected movie or TV program. Considering that some users don't like to talk, they only use gestures of static pictures to realize TV control without combining voice. Voice combined with gestures of static pictures realizes the control of the TV. After the voice wakes up, each frame of pictures is input and a g...

AI Technical Summary

Problems solved by technology

[0005] 1. After the voice wakes up, if each frame of picture is input, there will be a gesture action output, and there may be multiple trigger gesture actions, such as the ok gesture above, if the video continues to be collected and the picture is continuously input, there will be multiple ok gestures Output, which may trigger multiple action outputs;

[0006] 2. At the same time, gesture output based on static pictures cannot meet the precise need to trigger two consecutive output of the same gesture action. For example, you want to trigger two consecutive page turning operations through gestures;

[0007] 3. Gesture output based on static pictures, no fault-tolerant processing when outputting gesture actions, once a gesture is recognized incorrectly and recognized as another gesture, it will be triggered by mistake

[0008] 4. Usually this kind of technology starts to wake up from the voice, and the gesture recognition is turned on. If the algorithm does not limit the scene, there may be a possibility of triggering the output of gesture results, especially the problem of misrecognition of pictures. For example, after the current voice wakes up, It is no longer in the list selection state, but the action of turning the page is triggered (because of misrecognition of the picture or unintentional gesture operation)

[0009] If the TV is controlled by gestures combined with static pictures, and voice is not combined, this usually means that after the TV is turned on, the camera is called. One of these is the possibility of false triggering. Second, for the TV, it improves cpu usage, third, for users, continuous collection of pictures, personal privacy issues are not protected

Images