Living Face Verification via Multi-CNNs

Abstract

In face verification applications, precision rate and identifying liveness are two key factors. Traditional methods usually recognize global faces and can not gain good enough results when the faces are captured from different ages, or there are some interference factors, such as facial shade, etc. Besides, the false face attack will pose a great security risk. To solve the above problems, this study examines how to achieve reliable living face verification based on Multi-CNNs (convolutional neural networks) and Bayes probability. Participants are required to make several expressions in random orders and contents, to ensure their liveness. Then, an effective component-based method is proposed for face recognition, and synthesized multiple CNNs can help reflecting intensities of different components. Eventually, the similarity between faces is calculated by integrating the results of each CNN, with the help of Bayes probability. Comparative experiments demonstrate that our algorithm outperforms traditional methods in face recognition accuracy. Moreover, our algorithm has unique preponderance in that it can verify the liveness of users, which can achieve higher security.

A Survey of Face Recognition based on Convolutional Neural Network

“…The highest level of accuracy is achieved by combining ArcFace with W-KNN, which is 91.3% in the test dataset. [47]. Detection of facial components (eyes, eyebrows, nose, mouth) is done using the Active Shape Model (ASM).…”

A Hybrid Method for Traffic Flow Forecasting Using Multimodal Deep Learning

“…The most representative data-driven model is deep learning [10], which can automatically extract the relevant deep features of traffic data from multiple levels. Recently, deep learning has proven to be very successful in many areas, e.g., image, audio and natural language processing tasks since the breakthrough of Hinton et al [11], and these researches show that deep learning models have a superior or comparable performance with state-of-the-art methods in many fields [12][13][14][15]19,26]. Because traffic congestion process and traffic flow evolution are dynamic and nonlinear in nature, and deep learning model can learn the deep features of traffic data without prior knowledge.…”