Unconstrained Still/Video-Based Face Verification with Deep Convolutional Neural Networks

Deep learning features in facial identification and the likelihood ratio bound

In recent years, the score-based likelihood ratio (SLR) method for facial comparison has attracted considerable research attention. This method relies on the match scores that are calculated from the features obtained from facial recognition systems, deep learning based in particular. However, this concept has not been completely understood. Therefore, this study is aimed at investigating deep learning facial features, and the SLR levels of their match scores. We propose a new interpretation that the deep learning feature is a class characteristic. Based on a large-scale data set experiment, we present evidence that the log SLR value of deep learning features can reach 8 in some data sets. The study results imply that the SLR of deep learning features is a useful method for facial identification, especially when the suspected image is obtained via a CCTV camera.

A feature map aggregation network for unconstrained video face recognition

Unconstrained video face recognition is an extension of face recognition technology, and it is an indispensable part of intelligent security and criminal investigation systems. However, general face recognition technology cannot be directly applied to unconstrained video face recognition, because the video contains fewer frontal face image frames and a single image contains less face feature information. To address the above problems, this work proposes a Feature Map Aggregation Network (FMAN) to achieve unconstrained video face recognition by aggregating multiple face image frames. Specifically, an image group is used as the input of the feature extraction network to replace a single image to obtain a multi-channel feature map group. Then a quality perception module is proposed to obtain quality scores for feature maps and adaptively aggregate image features from image groups at the feature map level. Finally, extensive experiments are conducted on the challenging face recognition benchmarks YTF, IJB-A and COX to evaluate the proposed method, showing a significant increase in accuracy compared to the state-of-the-art.

Visual Feature Learning on Video Object and Human Action Detection: A Systematic Review

Video object and human action detection are applied in many fields, such as video surveillance, face recognition, etc. Video object detection includes object classification and object location within the frame. Human action recognition is the detection of human actions. Usually, video detection is more challenging than image detection, since video frames are often more blurry than images. Moreover, video detection often has other difficulties, such as video defocus, motion blur, part occlusion, etc. Nowadays, the video detection technology is able to implement real-time detection, or high-accurate detection of blurry video frames. In this paper, various video object and human action detection approaches are reviewed and discussed, many of them have performed state-of-the-art results. We mainly review and discuss the classic video detection methods with supervised learning. In addition, the frequently-used video object detection and human action recognition datasets are reviewed. Finally, a summarization of the video detection is represented, e.g., the video object and human action detection methods could be classified into frame-by-frame (frame-based) detection, extracting-key-frame detection and using-temporal-information detection; the methods of utilizing temporal information of adjacent video frames are mainly the optical flow method, Long Short-Term Memory and convolution among adjacent frames.

Application of Convolution Neural Network (CNN) Model Combined with Pyramid Algorithm in Aerobics Action Recognition

In order to realize high-accuracy recognition of aerobics actions, a highly applicable deep learning model and faster data processing methods are required. Therefore, it is a major difficulty in the field of research on aerobics action recognition. Based on this, this paper studies the application of the convolution neural network (CNN) model combined with the pyramid algorithm in aerobics action recognition. Firstly, the basic architecture of the convolution neural network model based on the pyramid algorithm is proposed. Combined with the application strategy of the common recognition model in aerobics action recognition, the traditional aerobics action capture information is processed. Through the characteristics of different aerobics actions, different accurate recognition is realized, and then, the error of the recognition model is evaluated. Secondly, the composite recognition function of the convolution neural network model in this application is constructed, and the common data layer effect recognition method is used in the optimization recognition. Aiming at the shortcomings of the composite recognition function, the pyramid algorithm is used to improve the convolution neural network recognition model by deep learning optimization. Finally, through the effectiveness comparison experiment, the results show that the convolution neural network model based on the pyramid algorithm is more efficient than the conventional recognition method in aerobics action recognition.

Construction and 3D Simulation of Virtual Animation Instant Network Communication System Based on Convolution Neural Networks

In recent years, great progress has been made in 3D simulation modeling of instant network communication system, such as the application of virtual reality technology and 3D virtual animation online modeling technology. Facing the increasing demand of different industries, how to build an instant network communication system for 3D virtual animation has become a research hotspot. On this basis, the construction method of fast instant network communication system based on convolutional neural network and fusion morphological 3D simulation model is studied. This paper analyzes the research status of instant network communication system. The experiment optimizes and improves the shortcomings of the current research hotspot of virtual animation instant network communication system and takes the morphological 3D simulation model fusion as the core for in-depth optimization. Finally, the experimental results show that the fusion morphological 3D simulation model can reconstruct the standard 3D virtual animation model according to different needs and can quickly model the optimization strategy according to the local differences of different animations. The response accuracy of the network communication system reaches 97.7%.

3D-Aided Dual-Agent GANs for Unconstrained Face Recognition

Synthesizing realistic profile faces is beneficial for more efficiently training deep pose-invariant models for large-scale unconstrained face recognition, by augmenting the number of samples with extreme poses and avoiding costly annotation work. However, learning from synthetic faces may not achieve the desired performance due to the discrepancy betwedistributions of the synthetic and real face images. To narrow this gap, we propose a Dual-Agent Generative Adversarial Network (DA-GAN) model, which can improve the realism of a face simulator's output using unlabeled real faces while preserving the identity information during the realism refinement. The dual agents are specially designed for distinguishing real versus fake and identities simultaneously. In particular, we employ an off-the-shelf 3D face model as a simulator to generate profile face images with varying poses. DA-GAN leverages a fully convolutional network as the generator to generate high-resolution images and an auto-encoder as the discriminator with the dual agents. Besides the novel architecture, we make several key modifications to the standard GAN to preserve pose, texture as well as identity, and stabilize the training process: (i) a pose perception loss; (ii) an identity perception loss; (iii) an adversarial loss with a boundary equilibrium regularization term. Experimental results show that DA-GAN not only achieves outstanding perceptual results but also significantly outperforms state-of-the-arts on the large-scale and challenging NIST IJB-A and CFP unconstrained face recognition benchmarks. In addition, the proposed DA-GAN is also a promising new approach for solving generic transfer learning problems more effectively. DA-GAN is the foundation of our winning entry to the NIST IJB-A face recognition competition in which we secured the $1^{st}$ places on the tracks of verification and identification.