Gesture recognition method, device, electronic device, and storage medium

1. A gesture recognition method, comprising:

sequentially performing a recognition process on each image from a target video using a preset recognition model of palm orientation to determine a probability of containing a palm image in each image and a palm normal vector corresponding to each image;

determining a group of target images from the target video based on the probability of containing the palm image in each image;

determining a target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images; and

training the preset recognition model of palm orientation by: obtaining a training data set, the training data set comprising a palm image, a non-palm image, and a palm normal vector corresponding to each palm image; and training an initial network model using the training data set to generate the preset recognition model of palm orientation.

2. The method of claim 1 , wherein determining the group of target images from the target video based on the probability of containing the palm image in each image comprises:

in response to detecting that the probability of containing the palm image in a first image is greater than or equal to a preset threshold, determining the first image as an image in the group of target images.

3. The method of claim 1 , wherein determining the target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images comprises:

determining an orientation of the palm in each target image based on the palm normal vector corresponding to each target image in the group of target images; and

determining the target gesture corresponding to the target video based on the orientation of the palm in each target image and an acquisition sequence of each target image.

4. The method of claim 1 , further comprising:

normalizing the palm normal vector of each palm image to obtain a unit normal vector of each palm image.

5. The method of claim 1 , further comprising:

in response to determining that the target gesture matches a first preset gesture of one or more preset gestures, controlling an electronic device based on a control command corresponding the first preset gesture.

6. An electronic device, comprising a processor and a memory, wherein the processor runs a program corresponding to an executable program code by reading the executable program code stored in the memory, such that the processor is configured to:

sequentially perform a recognition process on each image from a target video using a preset recognition model of palm orientation to determine a probability of containing a palm image in each image and a palm normal vector corresponding to each image;

determine a group of target images from the target video based on the probability of containing the palm image in each image; and

determine a target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images;

wherein the processor is further configured to train the preset recognition model of palm orientation by: obtaining a training data set, the training data set comprising a palm image, a non-palm image, and a palm normal vector corresponding to each palm image; and training an initial network model using the training data set to generate the preset recognition model of palm orientation.

7. The electronic device of claim 6 , wherein the processor is configured to determine the group of target images from the target video based on the probability of containing the palm image in each image by:

in response to detecting that the probability of containing the palm image in a first image is greater than or equal to a preset threshold, determining the first image as an image in the group of target images.

8. The electronic device of claim 6 , wherein the processor is configured to determine the target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images by:

determining an orientation of the palm in each target image based on the palm normal vector corresponding to each target image in the group of target images; and

determining the target gesture corresponding to the target video based on the orientation of the palm in each target image and an acquisition sequence of each target image.

9. The electronic device of claim 6 , wherein the processor is configured to:

normalize the palm normal vector of each palm image to obtain a unit normal vector of each palm image.

10. The electronic device of claim 6 , wherein the processor is further configured to:

in response to determining that the target gesture matches a first preset gesture of one or more preset gestures, control an electronic device based on a control command corresponding the first preset gesture.

11. A non-transitory computer readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the gesture recognition method is implemented, the method comprising:

sequentially performing a recognition process on each image from a target video using a preset recognition model of palm orientation to determine a probability of containing a palm image in each image and a palm normal vector corresponding to each image;

determining a group of target images from the target video based on the probability of containing the palm image in each image;

determining a target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images;

training the preset recognition model of palm orientation by: obtaining a training data set, the training data set comprising a palm image, a non-palm image, and a palm normal vector corresponding to each palm image; and training an initial network model using the training data set to generate the preset recognition model of palm orientation.

12. The non-transitory computer readable storage medium of claim 11 , wherein determining the group of target images from the target video based on the probability of containing the palm image in each image comprises:

in response to detecting that the probability of containing the palm image in a first image is greater than or equal to a preset threshold, determining the first image as an image in the group of target images.

13. The non-transitory computer readable storage medium of claim 11 , wherein determining the target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images comprises:

determining an orientation of the palm in each target image based on the palm normal vector corresponding to each target image in the group of target images; and

determining the target gesture corresponding to the target video based on the orientation of the palm in each target image and an acquisition sequence of each target image.

14. The non-transitory computer readable storage medium of claim 11 , wherein the method further comprises:

normalizing the palm normal vector of each palm image to obtain a unit normal vector of each palm image.

15. The non-transitory computer readable storage medium of claim 11 , wherein the method further comprises:

in response to determining that the target gesture matches a first preset gesture of one or more preset gestures, controlling an electronic device based on a control command corresponding the first preset gesture.