Face Recognition Algorithm Based on Multiscale Feature Fusion Network

Abstract

A face recognition model based on a multiscale feature fusion network is constructed, aiming to make full use of the characteristics of face and to improve the accuracy of face recognition. In addition, three different scale networks are designed to extract global features of faces. Multiscale cross-layer bilinear features of multiple networks are integrated via introducing a hierarchical bilinear pooling layer. By capturing some of the feature relationships between different levels, the model's ability to extract and distinguish subtle facial features is enhanced. Simultaneously, this study uses layer-by-layer deconvolution to fuse multilayer feature information, to solve the problem of losing some key features when extracting features from multilayer convolutional layers and pooled layers. The experimental results show that compared with the recognition accuracy of traditional algorithms, the recognition accuracy of the algorithm on Yale, AR, and ORL face databases is significantly improved.

Facial and speech Emotional Recognition based on Improved Deep Model

“…In expansion, the CNN-CLBP calculation can recognize 97% of cheerful expressions and shocked expressions, but the misidentification rate of pitiful expressions is 22.54%. In [10], a confront acknowledgment show is built based on a multi-scale highlight half breed organize, with the point of completely utilizing confront highlights and moving forward confront acknowledgment precision. In expansion, three distinctive scale systems are planned to extricate the worldwide highlights of faces.…”

Research on Yolo-based target recognition detection

“…Region suggestion is a multi-scale extraction of feature information from images by sliding windows of different sizes to find regions where a given object may be located. Feature extraction [6] is the conversion of images in candidate regions into feature vectors using manual feature extraction, commonly used methods such as local binary pattern features, gradient histogram features, etc. Classification regression is to predict the class of the target in the candidate region using a pre-trained classifier.…”