Dynamic and static gesture recognition method based on knowledge distillation

The invention provides a dynamic and static gesture recognition method based on knowledge distillation, and relates to the technical field of image recognition, and the method comprises the steps: carrying out the key frame extraction of a dynamic gesture, carrying out the extraction of the features of the dynamic gesture and the features of a static gesture through a three-dimensional convolution network and a two-dimensional convolution network, and finally, carrying out the recognition of the dynamic and static gestures. And carrying out distillation learning through a knowledge distillation learning framework composed of a teacher network and a student network. According to the method, the key frames are extracted, so that the problems of large distance between the key frames, serious insufficiency of key information, poor extraction precision and the like can be effectively solved, and the recognition accuracy and stability are improved; features of dynamic gestures and static gestures are extracted resp