Face detection with boosted Gaussian features

Detecting faces in images is a key step in numerous computer vision applications, such as face recognition or facial expression analysis. Automatic face detection is a difficult task because of the large face intra-class variability which is due to the important influence of the environmental conditions on the face appearance. We propose new features based on anisotropic Gaussian filters for detecting frontal faces in complex images. The performances of our face detector based on these new features have been evaluated on reference test sets, and clearly show improvements compared to the state-of-the-art.     

Face recognition based on dictionary learning and subspace learning

Dictionary learning plays an important role in sparse representation based face recognition. Many dictionary learning algorithms have been successfully applied to face recognition. However, for corrupted data because of noise or face variations (e.g. occlusion and large pose variation), their performances decline due to the disparity between domains. In this paper, we propose a face recognition algorithm based on dictionary learning and subspace learning (DLSL). In DLSL, a new subspace learning algorithm (SL) is proposed by using sparse constraint, low-rank technology and our label relaxation model to reduce the disparity between domains. Meanwhile, we propose a high-performance dictionary learning algorithm (HPDL) by constructing the embedding term, non-local self-similarity term, and time complexity drop term. In the obtained subspace, we use HPDL to classify these mapped test samples. DLSL is compared with other 28 algorithms on FRGC, LFW, CVL, Yale B and AR face databases. Experimental results show that DLSL achieves better performance than those 28 algorithms, including many state-of-the-art algorithms, such as recurrent regression neural network (RRNN), multimodal deep face recognition (MDFR) and projective low-rank representation (PLR).     

Tensor-Based AAM with Continuous Variation Estimation: Application to Variation-Robust Face Recognition

The active appearance model (AAM) is a well-known model that can represent a non-rigid object effectively. However, the fitting result is often unsatisfactory when an input image deviates from the training images due to its fixed shape and appearance model. To obtain more robust AAM fitting, we propose a tensor-based AAM that can handle a variety of subjects, poses, expressions, and illuminations in the tensor algebra framework, which consists of an image tensor and a model tensor. The image tensor estimates image variations such as pose, expression, and illumination of the input image using two different variation estimation techniques: discrete and continuous variation estimation. The model tensor generates variation-specific AAM basis vectors from the estimated image variations, which leads to more accurate fitting results. To validate the usefulness of the tensor-based AAM, we performed variation-robust face recognition using the tensor-based AAM fitting results. To do, we propose indirect AAM feature transformation. Experimental results show that tensor-based AAM with continuous variation estimation outperforms that with discrete variation estimation and conventional AAM in terms of the average fitting error and the face recognition rate.     

Towards more efficient and flexible face image deblurring using robust salient face landmark detection

Recent years have witnessed great progress in image deblurring. However, as an important application case, the deblurring of face images has not been well studied. Most existing face deblurring methods rely on exemplar set construction and candidate matching, which not only cost much computation time but also are vulnerable to possible complex or exaggerated face variations. To address the aforementioned problems, we propose a novel face deblurring method by integrating classical L ₀ deblurring approach with face landmark detection. A carefully tailored landmark detector is used to detect the main face contours. Then the detected contours are used as salient edges to guide the blind image deconvolution. Extensive experimental results demonstrate that the proposed method can better handle various complex face poses, shapes and expressions while greatly reducing computation time, as compared with existing state-of-the-art approaches.     

A robust and efficient convolutional deep learning framework for age-invariant face recognition

Extensive research has been carried out in the past on face recognition, face detection, and age estimation. However, age-invariant face recognition (AIFR) has not been explored that thoroughly. The facial appearance of a person changes considerably over time that results in introducing significant intraclass variations, which makes AIFR a very challenging task. Most of the face recognition studies that have addressed the ageing problem in the past have employed complex models and handcrafted features with strong parametric assumptions. In this work, we propose a novel deep learning framework that extracts age-invariant and generalized features from facial images of the subjects. The proposed model trained on facial images from a minor part (20–30%) of lifespan of subjects correctly identifies them throughout their lifespan. A variety of pretrained 2D convolutional neural networks are compared in terms of accuracy, time, and computational complexity to select the most suitable network for AIFR. Extensive experimental results are carried out on the popular and challenging face and gesture recognition network ageing dataset. The proposed method achieves promising results and outperforms the state-of-the-art AIFR models by achieving an accuracy of 99%, which proves the effectiveness of deep learning in facial ageing research.     

Class dependent factor analysis and its application to face recognition

We propose a class dependent factor analysis model (CDFA) which can be used in the general face recognition task under certain variations. The model utilizes the class information in a supervised manner to define a separate manifold for each class. Inside each manifold, a mixture of Gaussians is designated to handle the variation. The proposed model learns the system parameters in a probabilistic framework, allowing a Bayesian decision model. A manifold embedding technique is incorporated to handle the nonlinearity introduced by the variation; hence, a novel connection between manifold learning and probabilistic generative models is proposed. CDFA has better recognition accuracy and scalability over a classical factor analysis model. Experimental evaluations on the face recognition under changing illumination conditions and facial expressions indicate the ability of the proposed model to handle different types of variation. The achieved recognition rates are comparable to the state-of-art results, while it is also shown that the recognition rate does not decrease critically as the number of gallery identities increases.     

Learning Structure Models with Context Information for Visual Tracking

Tracking objects that undergo abrupt appearance changes and heavy occlusions is a challenging problem which conventional tracking methods can barely handle.To address the problem, we propose an online structure learning algorithm that contains three layers: an object is represented by a mixture of online structure models (OSMs) which are learnt from block-based online random forest classifiers (BORFs).BORFs are able to handle occlusion problems since they model local appearances of the target.To further improve the tracking accuracy and reliability, the algorithm utilizes mixture relational models (MRMs) as multi-mode context information to integrate BORFs into OSMs.Furthermore, the mixture construction of OSMs can avoid over-fitting effectively and is more flexible to describe targets.Fusing BORFs with MRMs, OSMs capture the discriminative parts of the target, which guarantees the reliability and robustness of our tracker.In addition, OSMs incorporate with block occlusion reasoning to update our BORFs and MRMs, which can deal with appearance changes and drifting problems effectively.Experiments on challenging videos show that the proposed tracker performs better than several state-of-the-art algorithms.     

Robust online appearance models for visual tracking

We propose a framework for learning robust, adaptive, appearance models to be used for motion-based tracking of natural objects. The model adapts to slowly changing appearance, and it maintains a natural measure of the stability of the observed image structure during tracking. By identifying stable properties of appearance, we can weight them more heavily for motion estimation, while less stable properties can be proportionately downweighted. The appearance model involves a mixture of stable image structure, learned over long time courses, along with two-frame motion information and an outlier process. An online EM-algorithm is used to adapt the appearance model parameters over time. An implementation of this approach is developed for an appearance model based on the filter responses from a steerable pyramid. This model is used in a motion-based tracking algorithm to provide robustness in the face of image outliers, such as those caused by occlusions, while adapting to natural changes in appearance such as those due to facial expressions or variations in 3D pose.     

基于性别和年龄因子分析的鲁棒性人脸表情识别

针对非可控环境下人脸表情识别面临的诸如种族、性别和年龄等因子变化问题,提出一种基于深度条件随机森林的鲁棒性人脸表情识别方法.与传统的单任务人脸表情识别方法不同,设计了一种以人脸表情识别为主,人脸性别和年龄属性识别为辅的多任务识别模型.在研究中发现,人脸性别和年龄等属性对人脸表情识别有一定的影响,为了捕获它们之间的关系,提出一种基于人脸性别和年龄双属性的深度条件随机森林人脸表情识别方法.在特征提取阶段,采用多示例注意力机制进行人脸特征提取以便去除诸如光照、遮挡和低分辨率等变化问题;在人脸表情识别阶段,根据人脸性别和年龄双属性因子,采用多条件随机森林方法进行人脸表情识别.在公开的CK+,ExpW,RAF-DB,AffectNet人脸表情数据库上进行了大量实验:在经典的CK+人脸库上达到99%识别率,在具有挑战性的自然场景库(ExpW,RAF-DB,AffectNet组合库)上达到70.52%的识别率.实验结果表明:与其他方法相比具有先进性,对自然场景中的遮挡、噪声和分辨率变化具有一定的鲁棒性.     

Scene Aware Detection and Block Assignment Tracking in crowded scenes

How far can human detection and tracking go in real world crowded scenes? Many algorithms often fail in such scenes due to frequent and severe occlusions as well as viewpoint changes. In order to handle these difficulties, we propose Scene Aware Detection (SAD) and Block Assignment Tracking (BAT) that incorporate with some available scene models (e.g. background, layout, ground plane and camera models). The SAD is proposed for accurate detection through utilizing 1) camera model to deal with viewpoint changes by rectifying sub-images, 2) a structural filter approach to handle occlusions based on a feature sharing mechanism in which a three-level hierarchical structure is built for humans, and 3) foregrounds for pruning negative and false positive samples and merging intermediate detection results. Many detection or appearance based tracking systems are prone to errors in occluded scenes because of failures of detectors and interactions of multiple objects. Differently, the BAT formulates tracking as a block assignment process, where blocks with the same label form the appearance of one object. In the BAT, we model objects on two levels, one is the ensemble level to measure how it is like an object by discriminative models, and the other one is the block level to measure how it is like a target object by appearance and motion models. The main advantage of BAT is that it can track an object even when all the part detectors fail as long as the object has assigned blocks. Extensive experiments in many challenging real world scenes demonstrate the efficiency and effectiveness of our approach.     

基于多部位多示例学习的人体检测

基于部位的检测方法能处理多姿态及部分遮挡的人体检测,多示例学习能有效处理图像的多义性,被广泛应用于图像检索与场景理解中。文中提出一种基于多示例学习的多部位人体检测方法。首先,根据人体生理结构将图像分割成若干区域,每个区域包含多个示例,利用AdaBoost多示例学习算法来训练部位检测器。然后利用各部位检测器对训练样本进行测试得到其响应值,从而将训练样本转化为部位响应值组成的特征向量。再用SVM方法对这些向量进行学习,得到最终的部位组合分类器。在INRIA数据集上的实验结果表明该方法能改进单示例学习的检测性能,同时评价3种不同的部位划分及其对检测性能的影响。     

Recognizing expressions from face and body gesture by temporal normalized motion and appearance features

Recently, recognizing affects from both face and body gestures attracts more attentions. However, it still lacks of efficient and effective features to describe the dynamics of face and gestures for real-time automatic affect recognition. In this paper, we combine both local motion and appearance feature in a novel framework to model the temporal dynamics of face and body gesture. The proposed framework employs MHI-HOG and Image-HOG features through temporal normalization or bag of words to capture motion and appearance information. The MHI-HOG stands for Histogram of Oriented Gradients (HOG) on the Motion History Image (MHI). It captures motion direction and speed of a region of interest as an expression evolves over the time. The Image-HOG captures the appearance information of the corresponding region of interest. The temporal normalization method explicitly solves the time resolution issue in the video-based affect recognition. To implicitly model local temporal dynamics of an expression, we further propose a bag of words (BOW) based representation for both MHI-HOG and Image-HOG features. Experimental results demonstrate promising performance as compared with the state-of-the-art. Significant improvement of recognition accuracy is achieved as compared with the frame-based approach that does not consider the underlying temporal dynamics.     

Multi-view facial landmark detector learned by the Structured Output SVM

We propose a real-time multi-view landmark detector based on Deformable Part Models (DPM). The detector is composed of a mixture of tree based DPMs, each component describing landmark configurations in a specific range of viewing angles. The usage of view specific DPMs allows to capture a large range of poses and to deal with the problem of self-occlusions. Parameters of the detector are learned from annotated examples by the Structured Output Support Vector Machines algorithm. The learning objective is directly related to the performance measure used for detector evaluation. The tree based DPM allows to find a globally optimal landmark configuration by the dynamic programming. We propose a coarse-to-fine search strategy which allows real-time processing by the dynamic programming also on high resolution images. Empirical evaluation on “in the wild” images shows that the proposed detector is competitive with the state-of-the-art methods in terms of speed and accuracy yet it keeps the guarantee of finding a globally optimal estimate in contrast to other methods.     

Video-to-video face authentication system robust to pose variations

High-quality still-to-still (image-to-image) face authentication has shown success under controlled conditions in many safety applications. However, video-to-video face authentication is still challenging due to appearance variations caused by pose changes. In this paper, we propose a video-to-video face authentication system that is robust to pose variations by making use of synthesized frontal face appearance that contains both texture and shape information. To obtain the appearance, we first reconstruct 3D face shape from face feature points detected from the video using active shape model (ASM). Conventional ASM algorithms cannot handle large pose variations and fast head movement exhibited in video sequences. To address these problems, we present a novel prediction-assisted approach that is capable of providing an accurate shape initiation as well as automatically switching on multi-view models for ASM. Then we can generate frontal shape mesh from the reconstructed 3D face shape. Based on the mesh, we synthesize frontal face appearance with the ASM-detected faces in video. For authentication, in order to match the synthesized appearances of enrollment and probe, we propose a 2-directional 2-dimensional client specific fisher’s linear discriminant algorithm. The proposed algorithm is a variant of fisher’s linear discriminant (FLD) and directly computes eigenvectors of image scatter matrices in row and column direction without matrix-to-vector conversion. In experiments, our authentication system is compared with the other state-of-art approaches on public face database and our face database. The results show that our system demonstrates a higher authentication accuracy and pose-robust performance.     

Adaptive multi-cue tracking by online appearance learning

This paper proposes a multi-cue based appearance learning algorithm for object tracking. In each frame, the target object is represented by different cues in the image-as-matrix form. This representation can describe the target from different perspectives and can preserve the spatial correlation information inside the target region. Based on these cues, multiple appearance models are learned online by bilinear subspace analysis to account for the target appearance variations over time. Tracking is formulated within the Bayesian inference framework, in which the observation model is constructed by fusing all the learned appearance models. The combination of online appearance modeling and weight update of each appearance model can adapt our tracking algorithm to both the target and background changes. We test our algorithm on a variety of challenging sequences by tracking car, face, pedestrian, and so on. Experimental results and comparisons to several state-of-the-art methods show improved tracking performance.     

Weakly Supervised Localization and Learning with Generic Knowledge

Learning a new object class from cluttered training images is very challenging when the location of object instances is unknown, i.e. in a weakly supervised setting. Many previous works require objects covering a large portion of the images. We present a novel approach that can cope with extensive clutter as well as large scale and appearance variations between object instances. To make this possible we exploit generic knowledge learned beforehand from images of other classes for which location annotation is available. Generic knowledge facilitates learning any new class from weakly supervised images, because it reduces the uncertainty in the location of its object instances. We propose a conditional random field that starts from generic knowledge and then progressively adapts to the new class. Our approach simultaneously localizes object instances while learning an appearance model specific for the class. We demonstrate this on several datasets, including the very challenging Pascal VOC 2007. Furthermore, our method allows training any state-of-the-art object detector in a weakly supervised fashion, although it would normally require object location annotations.     

稀疏综合字典学习的小样本人脸识别

传统以字典学习为基础的小样本人脸识别方法存在字典低辨别性、弱鲁棒性等缺点,对此,本文提出稀疏综合字典学习模型。该模型有效利用和生成人脸变化,以镜像原理及Fisher准则扩充训练样本多样性,通过构造混合特色字典、扩充干扰字典以及低秩字典原子,提取不同类别数据之间的共性、特殊性和异常情况,从而提高算法识别率以及对表情变化、姿态变化、遮挡等异常情况的处理能力。在AR、YALEB、LFW等人脸数据库进行仿真实验,实验结果验证了算法的有效性和可行性。     

OptiFuzz: a robust illumination invariant face recognition system and its implementation

Vision-based human face detection and recognition are widely used and have been shown to be effective in normal illumination conditions. Under severe illumination conditions, however, it is very challenging. In this paper, we address the effect of illumination on the face detection and the face recognition problem by introducing a novel illumination invariant method, called OptiFuzz. It is an optimized fuzzy-based illumination invariant method to solve the effect of illumination for photometric-based human face recognition. The rule of the Fuzzy Inference System is optimized by using a genetic algorithm. The Fuzzy’s output controls an illumination invariant model that is extended from Land’s reflectance model. We test our method by using Yale B Extended and CAS-PEAL face databases to represent the offline experiments, and several videos are recorded at our campus to represent the online indoor and outdoor experiments. Viola–Jones face detector and mutual subspace method are employed to handle the online face detection and face recognition experiments. Based on the experimental results, we can show that our algorithm outperforms the existing and the state-of-the-art methods in recognizing a specific person under variable lighting conditions with a significantly improved computation time. Other than that, using illumination invariant images is also effective in improving the face detection performance.     

Network representation learning based on community-aware and adaptive random walk for overlapping community detection

The Network representation learning methods based on random walk aim to learn a low-dimensional embedding vector for each node in a network by randomly traversing the network to capture the features of nodes and edges, which is beneficial to many downstream machine learning tasks such as community detection. Most of the existing random-walk-based network representation learning algorithms emphasize the neighborhood of nodes but ignore the communities they may form and apply the same random walk strategy to all nodes without distinguishing the characteristics of different nodes. In addition, it is time-consuming to determine the most suitable random walk parameters for a given network. In this paper, we propose a novel overlapping community detection algorithm based on network representation learning which integrates community information into embedding vectors to improve the cohesion degree of similar nodes in the embedding space. First, a node-centrality-based walk strategy is designed to determine the parameters of random walk automatically to avoid the time-consuming manual selection. Second, two community-aware random walk strategies for high and low degree nodes are developed to capture the characteristics of the community centers and boundaries. The experimental results on the synthesized and real-world datasets demonstrate the effectiveness and efficiency of our algorithm on overlapping community detection compared with the state-of-the-art algorithms