Multi-view Face Research Articles

Adaptive Fusion of Multi-Biometrics for Human Identification in Video

The purpose of multiple biometric fusion is to improve the recognition performance by utilizing their complementary. In this paper, the feature fusion recognition method of multi-view face and gait in video is studied, and a adaptive decision fusion method is proposed. The results showed that the adaptive fusion features carry the most discriminating power compared to any individual biometric and other static fusion rules like Max and Sum.

MiLDA: A graph embedding approach to multi-view face recognition

In a vast number of real-world face recognition applications, gallery and probe image sets are captured from different scenarios. For such multi-view data, face recognition systems often perform poorly. To tackle this problem, in this paper we propose a graph embedding framework, which can project the multi-view data into a common subspace of higher discriminability between classes. This framework can be readily utilized to extend classical dimensionality reduction methods to multi-view scenarios. Hence, by utilizing the framework for multi-view face recognition, we propose multi-view linear discriminant analysis (MiLDA). We also empirically demonstrate that, for several distinct multi-view face recognition scenarios, MiLDA has an excellent performance and outperforms many popular approaches.

Multi-View Face Recognition Based on BP Neural Network and KPCA

This paper introduces the research background of computer face recognition technology, and puts forward a method of using kernel principal component analysis (KPCA) method and improved BP neural network methods for analysis and identification of multi view face images. The experimental results show that this algorithm is both effective and accurate. It achieved a higher recognition rate and excellent resistance to noise.

Multi-view Face Analysis Based on Gabor Features

Facial analysis has attracted much attention in the technology for human-machine interface. Different methods of classification based on sparse representation and Gabor kernels have been widely applied in the fields of facial analysis. However, most of these methods treat face from a whole view standpoint. In terms of the importance of different facial views, in this paper, we present multi-view face analysis based on sparse representation and Gabor wavelet coefficients. To evaluate the performance, we conduct face analysis experiments including Face Recognition (FR) and Face Expression Recognition (FER) on JAFFE database. Experiments are conducted from two parts: (1) Face images are divided into three facial parts which are forehead, eye and mouth. (2) Face images are divided into 8 parts by the orientation of Gabor kernels. Experimental results demonstrate that the proposed methods can significantly boost the performance and perform better than the other methods.

Appearance-based multimodal human tracking and identification for healthcare in the digital home.

There is an urgent need for intelligent home surveillance systems to provide home security, monitor health conditions, and detect emergencies of family members. One of the fundamental problems to realize the power of these intelligent services is how to detect, track, and identify people at home. Compared to RFID tags that need to be worn all the time, vision-based sensors provide a natural and nonintrusive solution. Observing that body appearance and body build, as well as face, provide valuable cues for human identification, we model and record multi-view faces, full-body colors and shapes of family members in an appearance database by using two Kinects located at a home's entrance. Then the Kinects and another set of color cameras installed in other parts of the house are used to detect, track, and identify people by matching the captured color images with the registered templates in the appearance database. People are detected and tracked by multisensor fusion (Kinects and color cameras) using a Kalman filter that can handle duplicate or partial measurements. People are identified by multimodal fusion (face, body appearance, and silhouette) using a track-based majority voting. Moreover, the appearance-based human detection, tracking, and identification modules can cooperate seamlessly and benefit from each other. Experimental results show the effectiveness of the human tracking across multiple sensors and human identification considering the information of multi-view faces, full-body clothes, and silhouettes. The proposed home surveillance system can be applied to domestic applications in digital home security and intelligent healthcare.

Multi-View Face Detection Based on Multi-Features AdaBoost Collaborative Learning Algorithm

In View of the Multi-View Face Detection Problem under Complex Background, an Improved Face Detection Method Based on Multi-Features Boosting Collaborative Learning Algorithm Integrating Local Binary Pattern (LBP) is Presented. Firstly, Facial Skin Color Information is Used to Exclude most of the Background Regions. then, Haar-like Feature and LBP Feature are Extracted from Facial Candidate Regions and Inputted into a Modified Adaboost Algorithm to Obtain a Strong Classifier. Lastly, in Order to Improve the Detection Speed, Pyramid Classifier System Structure is Adopted to Determine the Face. the Experimental Results on CMU Standard Test Set and Life Photos, the Proposed Method has Achieved the Rapid Detection of Multi-View Face Image.

Comparison of Multi-View Face Recognition using DCT and Hybrid DWT of Score Fusion under Uncontrolled Illumination Variation

A facial recognition system is one of the biometric applications for automatically identifying or verifying a person from a digital image or a video frame from a video source. Now a day’s face recognition system is recognize the face using multiple-views of faces, for detecting each view of face such as left, right, front, top, and bottom. The proposed face recognition system consists of a novel illuminationinsensitive preprocessing method, a hybrid discrete Wavelet Transform (HDWT) for feature extraction and Density based Score Fusion Technique for matching. First, in the preprocessing stage different stages are followed like Gamma Correction, DOG(Difference of Gaussion)filtering and contrast equalization which transforms, a face image into an illumination-insensitive image, Then, for feature extraction of complementary classifiers, multiple face models based upon HDWT are applied. To do the feature extraction this system uses DCT Wavelet transform to generate the feature vectors of the query and database images. Euclidean distance is used as similarity measure to compare the image. Finally, to combine scores from multiple complementary classifiers, a log likelihood ratio-based score fusion scheme is applied. The proposed system uses the ORL face database, and compare with DCT method. Using ORL database the Hybrid wavelet has achieves smaller size feature vectors which is performing better with 98.55% accuracy and less computation time require for recognizing image which is 19.0645 msec to 24.7653 msec .Which is less as compare to DCT method, DCT method achieves 96.45% accuracy and computational time 26.2349 msec to 33.0234 msec which is more as compare to proposed method..

Multiview face recognition based on multilinear decomposition and pose manifold

One major challenge encountered in face recognition is how to handle the wide pose variation and in-depth rotations of head. A multiview face recognition method is proposed in this study that addresses this challenge based on multilinear decomposition approach and pose subspace. In order to preserve the pose manifold geometry among different individuals in pose subspace, a pose-biased distance measure is proposed. In addition, as one of the impediments in manifold-based methods is the lack of sufficient data, a new half-ellipsoid-based pose generation method is presented. For performance evaluation of the proposed multiview face recognition method, three different experiments are run on three famous face datasets. The obtained recognition accuracy and the cumulative match characteristic curves confirm the effectiveness of the proposed method in wide pose variation, even with limited number of training poses.

Robust face recognition from multi-view videos.

Multiview face recognition has become an active research area in the last few years. In this paper, we present an approach for video-based face recognition in camera networks. Our goal is to handle pose variations by exploiting the redundancy in the multiview video data. However, unlike traditional approaches that explicitly estimate the pose of the face, we propose a novel feature for robust face recognition in the presence of diffuse lighting and pose variations. The proposed feature is developed using the spherical harmonic representation of the face texture-mapped onto a sphere; the texture map itself is generated by back-projecting the multiview video data. Video plays an important role in this scenario. First, it provides an automatic and efficient way for feature extraction. Second, the data redundancy renders the recognition algorithm more robust. We measure the similarity between feature sets from different videos using the reproducing kernel Hilbert space. We demonstrate that the proposed approach outperforms traditional algorithms on a multiview video database.

A Novel Multi-view Face Detection Method Based on Improved Real Adaboost Algorithm

A Novel Multi-view Face Detection Method Based on Improved Real Adaboost Algorithm

Multiview Face Detection and Registration Requiring Minimal Manual Intervention

Most face recognition systems require faces to be detected and localized a priori. In this paper, an approach to simultaneously detect and localize multiple faces having arbitrary views and different scales is proposed. The main contribution of this paper is the introduction of a face constellation, which enables multiview face detection and localization. In contrast to other multiview approaches that require many manually labeled images for training, the proposed face constellation requires only a single reference image of a face containing two manually indicated reference points for initialization. Subsequent training face images from arbitrary views are automatically added to the constellation (registered to the reference image) based on finding the correspondences between distinctive local features. Thus, the key advantage of the proposed scheme is the minimal manual intervention required to train the face constellation. We also propose an approach to identify distinctive correspondence points between pairs of face images in the presence of a large amount of false matches. To detect and localize multiple faces with arbitrary views, we then propose a probabilistic classifier-based formulation to evaluate whether a local feature cluster corresponds to a face. Experimental results conducted on the FERET, CMU, and FDDB datasets show that our proposed approach has better performance compared to the state-of-the-art approaches for detecting faces with arbitrary pose.

Enhanced long-range personal identification based on multimodal information of human features

This work presents an enhanced long-range personal identification scheme using multimodal information of human features. Multimodal information includes multiview face detection, height measurement and face recognition. Multiview faces are estimated by collecting five face databases that correspond to left, half-left, right, half-right, and frontal faces, respectively. The sequences of parameters based on four detectors are also designed to determine the face direction. The detectors use the head-shoulder region, frontal face, profile face, and eyes detector respectively. In addition to determining when individuals enter the monitoring area, the multiview face detection module also describes the detected face direction. This result allows the identification system to select the face database of a specific direction to identify subsequent faces. Additionally, the height measurement module estimates individual height by calculating the vanishing points and lines. The module concept is based on single-view metrology. The measured information further refines the face database selected by multiview face detection and minimizes the candidates for face identification. Importantly, the proposed method integrates the multimodal information based on face direction, height and face features to refine the database and analyzes the information to determine the identity of a person. In this work, images from a monitoring area 5.6 m away from a camera are captured using an inexpensive digital web camera. The experimental results show that the proposed method can improve the accuracy rate by more than 21 % in contrast with the baselines and correspondingly demonstrates the effectiveness of the proposed idea.

Face Detection using Color based Segmentation and Edge Detection

The increasing use of computer vision in security in place of humans led many to research the problem of face detection in images. The problem is not a petty one as the classification of a human face proves to challenging. Despite the many variations of a human face, features can still be found, given a certain context, which will uniquely identify a face. Early face-detection algorithms focused on the detection of frontal human faces, whereas this paper attempt to solve the more general and difficult problem of multi-view face detection. Face detection involves many research challenges such as scale, rotation, and pose and illumination variation. The techniques used for face detection have been researched for years and much progress has been suggested in literature. This paper proposes a new technique for detecting faces in color images using color model and edge detection. Face detection is used in as a part of a facial recognition system. It is also used in human computer interface, image database management and video surveillance. The results of this technique show that the proposed algorithm is good enough to detect the human face taken through video with accuracy. This paper is achieving high detection speed, high detection accuracy and reduces the false detecting rate.

Super-Resolution Method for Multiview Face Recognition From a Single Image Per Person Using Nonlinear Mappings on Coherent Features

In video surveillance, the face recognition usually aims at recognizing a nonfrontal low resolution face image from the gallery in which each person has only one high resolution frontal face image. Traditional face recognition approaches have several challenges, such as the difference of image resolution, pose variation and only one gallery image per person. This letter presents a regression based method that can successfully recognize the identity given all these difficulties. The nonlinear regression models from the specific nonfrontal low resolution image to frontal high resolution features are learnt by radial basis function in subspace built by canonical correlation analysis. Extensive experiments on benchmark database show the superiority of our method.

Multiview Face Recognition: From TensorFace to V-TensorFace and K-TensorFace

Face images under uncontrolled environments suffer from the changes of multiple factors such as camera view, illumination, expression, etc. Tensor analysis provides a way of analyzing the influence of different factors on facial variation. However, the TensorFace model creates a difficulty in representing the nonlinearity of view subspace. In this paper, to break this limitation, we present a view-manifold-based TensorFace (V-TensorFace), in which the latent view manifold preserves the local distances in the multiview face space. Moreover, a kernelized TensorFace (K-TensorFace) for multiview face recognition is proposed to preserve the structure of the latent manifold in the image space. Both methods provide a generative model that involves a continuous view manifold for unseen view representation. Most importantly, we propose a unified framework to generalize TensorFace, V-TensorFace, and K-TensorFace. Finally, an expectation-maximization like algorithm is developed to estimate the identity and view parameters iteratively for a face image of an unknown/unseen view. The experiment on the PIE database shows the effectiveness of the manifold construction method. Extensive comparison experiments on Weizmann and Oriental Face databases for multiview face recognition demonstrate the superiority of the proposed V- and K-TensorFace methods over the view-based principal component analysis and other state-of-the-art approaches for such purpose.

Joint dynamic sparse representation for multi-view face recognition

We consider the problem of automatically recognizing a human face from its multi-view images with unconstrained poses. We formulate the multi-view face recognition task as a joint sparse representation model and take advantage of the correlations among the multiple views for face recognition using a novel joint dynamic sparsity prior. The proposed joint dynamic sparsity prior promotes shared joint sparsity patterns among the multiple sparse representation vectors at class-level, while allowing distinct sparsity patterns at atom-level within each class to facilitate a flexible representation. Extensive experiments on the CMU Multi-PIE face database are conducted to verify the efficacy of the proposed method.

Multi-View Face Detection Based on Kernel Principal Component Analysis and Kernel Support Vector Techniques

Detecting faces across multiple views is more challenging than in a frontal view. To address this problem, an efficient approach is presented in this paper using a kernel machine based approach for learning such nonlinear mappings to provide effective view-based representation for multi-view face detection. In this paper Kernel Principal Component Analysis (KPCA) is used to project data into the view-subspaces then computed as view-based features. Multi-view face detection is performed by classifying each input image into face or non-face class, by using a two class Kernel Support Vector Classifier (KSVC). Experimental results demonstrate successful face detection over a wide range of facial variation in color, illumination conditions, position, scale, orientation, 3D pose, and expression in images from several photo collections.

Robust Facial Features Localization on Rotation Arbitrary Multi-View face in Complex Background

Focused on facial features localization on multi-view face arbitrarily rotated in plane, a novel detection algorithm based improved SVM is proposed. First, the face is located by the rotation invariant multi-view (RIMV) face detector and its pose in plane is corrected by rotation. After the searching ranges of the facial features are determined, the crossing detection method which uses the brow-eye and nose-mouth features and the improved SVM detectors trained by large scale multi-view facial features examples is adopted to find the candidate eye, nose and mouth regions,. Based on the fact that the window region with higher value in the SVM discriminant function is relatively closer to the object, and the same object tends to be repeatedly detected by near windows, the candidate eyes, nose and mouth regions are filtered and merged to refine their location on the multi-view face. Experiments show that the algorithm has very good accuracy and robustness to the facial features localization with expression and arbitrary face pose in complex background.

Robust Multi-Camera View Face Recognition

This paper presents multi-appearance fusion of principal component analysis (PCA) and generalization of linear discriminant analysis (LDA) for multi-camera view offline face recognition (verification) system. The generalization of LDA has been extended to establish correlations between the face classes in the transformed representation and this is called canonical covariate. The proposed system uses Gabor filter banks for characterization of facial features by spatial frequency, spatial locality and orientation to compensate to the variations of face instances occurred due to illumination, pose and facial expression changes. Convolution of Gabor filter bank to face images produces Gabor face representations with high-dimensional feature vectors. PC A and canonical covariate are then applied on the Gabor face representations to reduce the high-dimensional feature spaces into low-dimensional Gabor eigenfaces and Gabor canonical faces. Reduced eigenface vector and canonical face vector are fused together using weighted mean fusion rule. Finally, support vector machines have trained with augmented fused set of features and perform the recognition task. The system has been evaluated with UMIST face database consisting of multi-view faces. The experimental results demonstrate the efficiency and robustness of the proposed system for multi-view face images with high recognition rates. Complexity analysis of the proposed system is also presented at the end of the experimental results.

Distributed Classification of Multiple Observation Sets by Consensus

We consider the problem of distributed classification of multiple observations of the same object that are collected in an ad hoc network of vision sensors. Assuming that each sensor captures a different observation of the same object, the problem is to classify this object by distributed processing in the network. We present a graph-based problem formulation whose objective function captures the smoothness of candidate labels on the data manifold formed by the observations of the object. We design a distributed average consensus algorithm for estimating the unknown object class by computing the value of the smoothness objective function for different class hypotheses. It initially estimates the objective function locally based on the observation of each sensor. As the distributed consensus algorithm progresses, all observations are gradually taken into account in the estimation of the objective function. We illustrate the performance of the distributed classification algorithm for multiview face recognition in an ad hoc network of vision sensors. When the training set is sufficiently large, the simulation results show that the consensus classification decision is equivalent to the decision of a centralized system that has access to all observations.