Multi-feature fusion face detection based on attention mechanism

Summary: Aiming at the problems of insufficient feature information extraction, low accuracy, and easy to be deceived in conventional face live detection, aA multi-feature fusion face detection algorithm based on attention mechanism.First, forMake full use of the low-level features of the image, transform the original face image to obtain the Fourier spectrogram, and combine the local binary pattern map and the simplified Weber local description map as the input of the detection model.Secondly, put this 3types of images are entered into 3ResNetin tributaries.Then, an attention mechanism is introduced to represent the importance of different features for recognition, and the extracted features are weighted and fused.Finally, in the benchmark databaseCASIA-FASD