Abstract:With the application of deep convolutional neural networks, the performance of computer vision tasks has been improved to a new level. The construction of a deeper and more complex network allows the face recognition algorithm to obtain a higher accuracy, However, the disadvantages of large computation and storage costs of neural networks limit the further popularization of the algorithm. To solve this problem, we have studied the unified and efficient neural network face recognition algorithm under the condition of a single camera; we propose that the complete face recognition process consists of four tasks: face detection, in vivo detection, keypoint detection, and face verification; combining the key algorithms of these four tasks, we propose a unified network model based on a deep separable convolutional structure—UFaceNet. The model uses multisource data to carry out multitask joint training and uses the keypoint detection results to aid the learning of other tasks. It further introduces the attention mechanism through feature level clipping and alignment to ensure the accuracy of the model, using the shared convolutional layer network among tasks to reduce model calculations amount and realize network acceleration. The learning goal of multi-tasking implicitly increases the amount of training data and different data distribution, making it easier to learn the characteristics with generalization. The experimental results show that the UFaceNet model is better than other models in terms of calculation amount and number of parameters with higher efficiency, and some potential areas to be used.

Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks

End-to-End Spatial Transform Face Detection and Recognition

Cascaded Deep Convolutional Neural Network for Robust Face Alignment.

Improved Face Detection and Alignment using Cascade Deep Convolutional Network

Simultaneous Face Detection and Pose Estimation Using Convolutional Neural Network Cascade.

Joint Multi-View Face Alignment in the Wild

Face Alignment with Cascaded Bidirectional LSTM Neural Networks.

Multi-task Convolution Network for Face Alignment

Joint Learning for Face Alignment and Face Transfer with Depth Image

Face Alignment Algorithm Based on an Improved Cascaded Convolutional Neural Network

Towards efficient masked-face alignment via cascaded regression

Facial Landmark Detection Via Cascade Multi-Channel Convolutional Neural Network

Face Alignment by Deep Convolutional Network with Adaptive Learning Rate.

Adaptive Cascade Deep Convolutional Neural Networks for face alignment

Joint Face Detection and Initialization for Face Alignment.

Research and Implementation on Face Detection Approach Based on Cascaded Convolutional Neural Networks

Deep Adaptive Attention for Joint Facial Action Unit Detection and Face Alignment

Multi-face detection and alignment using multiple kernels

Region-Based Face Alignment With Convolution Neural Network Cascade

Joint 3d Face Reconstruction And Dense Face Alignment Via Deep Face Feature Alignment

UFaceNet: Research on Multi-Task Face Recognition Algorithm Based on CNN.