面向非受控场景的人脸图像正面化重建

目的 人脸正面化重建是当前视觉领域的热点问题。现有方法对于模型的训练数据具有较高的需求,如精确的输入输出图像配准、完备的人脸先验信息等。但该类数据采集成本较高,可应用的数据集规模较小,直接将现有方法应用于真实的非受控场景中往往难以取得理想表现。针对上述问题,提出了一种无图像配准和先验信息依赖的任意视角人脸图像正面化重建方法。方法 首先提出了一种具有双输入路径的人脸编码网络,分别用于学习输入人脸的视觉表征信息以及人脸的语义表征信息,两者联合构造出更加完备的人脸表征模型。随后建立了一种多类别表征融合的解码网络,通过以视觉表征为基础、以语义表征为引导的方式对两种表征信息进行融合,融合后的信息经过图像解码即可得到最终的正面化人脸图像重建结果。结果 首先在Multi-PIE(multi-pose, illumination and expression)数据集上与8种较先进方法进行了性能评估。定量和定性的实验结果表明,所提方法在客观指标以及视觉质量方面均优于对比方法。此外,相较于当前性能先进的基于光流的特征翘曲模型(flow-based feature warping model,FFWM)方法,本文方法能够节省79%的参数量和42%的计算操作数。进一步基于CASIA-WebFace(Institute of Automation, Chinese Academy of Sciences—WebFace)数据集对所提出方法在真实非受控场景中的表现进行了评估,识别精度超过现有方法10%以上。结论 本文提出的双层级表征集成推理网络,能够挖掘并联合人脸图像的底层视觉特征以及高层语义特征,充分利用图像自身信息,不仅以更低的计算复杂度取得了更优的视觉质量和身份识别精度,而且在非受控的场景下同样展现出了出色的泛化性能。     

人脸样貌独立判别的协作表情识别算法

为了降低样貌、姿态、眼镜以及表情定义不统一等因素对人脸表情识别的影响,提出一种人脸样貌独立判别的协作表情识别算法。首先,采用自动的人脸检测算法定位、对齐视频每帧的人脸区域,并从人脸视频序列中选择峰值表情的人脸;然后,采用峰值人脸与某个表情类内的所有人脸产生表情类内差异人脸信息,并通过计算峰值表情人脸与表情类内差异人脸的差异信息获得协作的表情表示;最终,采用基于稀疏的分类器与表情表示决定每个人脸表情的标签。采用欧美与亚洲人脸的数据库进行仿真实验,结果表明本算法获得了较好的表情识别准确率,对不同样貌、佩戴眼镜的人脸样本也具有较好的识别效果。     

人脸的层次化描述模型及识别研究

人脸自动识别是一个困难但有重要意义的工作。文中提出了一种基于人脸层次化描述的识别方法。该方法首先对人脸进行快速准确的特征定位及标准化,然后采用主元分析神经网络分别对定位的人脸及其特征区域进行最佳特征提取,从而得到人脸在低分辨率和较高分辨率上的两层特征描述用以识别,具有识别率高、特征数据量适中、可用于大量人像识别等特点。此方法在１３００幅人像上进行了测试,结果表明其在人脸转动、表情变化或入脸未经训练     

3D face authentication and recognition based on bilateral symmetry analysis

We present a novel and computationally fast method for automatic human face authentication. Taking a 3D triangular facial mesh as input, the approach first automatically extracts the bilateral symmetry plane of the facial surface. The intersection between the symmetry plane and the facial surface, namely the symmetry profile, is then computed. Using both the mean curvature plot of the facial surface and the curvature plot of the symmetry profile curve, three essential points of the nose on the symmetry profile are automatically extracted. The three essential points uniquely determine a Face Intrinsic Coordinate System (FICS). Different faces are aligned based on the FICS. The symmetry profile, together with two transverse profiles, composes a compact representation, called the SFC representation, of a 3D face surface. The face authentication and recognition steps are finally performed by comparing the SFC representations of the faces. The proposed method was tested on 382 face surfaces, which come from 166 individuals and cover a wide ethnic and age variety. The equal error rate (EER) of face authentication on scans with variable facial expressions is 10.8%. For scans with normal expression, the ERR is 0.8%.     

对称主分量分析及其在人脸识别中的应用

镜像对称性是人脸的一个直观显然的自然特性，有助于开发面向人脸图像的识别技术与算法，该文将在人脸识别中应用这一自然特性，提出一种新算法——对称主分量分析，该算法首先引入镜像变换，生成镜像样本；然后依据奇偶分解原理，生成镜象奇、偶对称样本，并分别进行KL展开，提取镜象奇／偶对称KL特征分量；最后，根据奇／偶对称KL特征分量在人脸中所占能量比例的不同以及对视角、旋转、光照等干扰的不同敏感程度进行特征选择，理论分析与实验证明，该算法巧妙地利用镜像样本，增强人脸识别：既扩大样本容量，显著提高识别率；又节省计算与存储开销，增强算法的实用性能。     

Leveraging Graph Cut’s Energy Function for Context Aware Facial Recognition in Indoor Environments

Context-aware facial recognition regards the recognition of faces in association with their respective environments. This concept is useful for the domestic robot which interacts with humans when performing specific functions in indoor environments. Deep learning models have been relevant in solving facial and place recognition challenges; however, they require the procurement of training images for optimal performance. Pre-trained models have also been offered to reduce training time significantly. Regardless, for classification tasks, custom data must be acquired to ensure that learning models are developed from other pre-trained models. This paper proposes a place recognition model that is inspired by the graph cut energy function, which is specifically designed for image segmentation. Common objects in the considered environment are identified and thereafter they are passed over to a graph cut inspired model for indoor environment classification. Additionally, faces in the considered environment are extracted and recognised. Finally, the developed model can recognise a face together with its environment. The strength of the proposed model lies in its ability to classify indoor environments without the usual training process(es). This approach differs from what is obtained in traditional deep learning models. The classification capability of the developed model was compared to state-of-the-art models and exhibited promising outcomes.     

Growing snakes: active contours for complex topologies

Snakes are active contours that minimize an energy function. In this paper we introduce a new kind of snakes, called growing snakes. These snakes are modeled as a set of particles connected by thin rods. Unlike the traditional snakes, growing snakes are automatically initialized. They start at the position where the gradient magnitude of an image is largest, and start to grow looking for zones of high gradient magnitude; simultaneously the associated energy function is minimized. Growing snakes can find contours with complex topology, describing holes, occlusions, separate objects and bifurcations. In a post-process the T-junctions are refined looking for the configuration with minimal energy. We also describe a technique that permits one to regularize the field of external forces that act on the Growing Snakes, which allow them to have good performance, even in the case of images with high levels of noise. Finally, we present results in synthetic and real images.     

Illuminant-Dependence of Von Kries Type Quotients

An expression-invariant 3D face recognition approach is presented. Our basic assumption is that facial expressions can be modelled as isometries of the facial surface. This allows to construct expression-invariant representations of faces using the bending-invariant canonical forms approach. The result is an efficient and accurate face recognition algorithm, robust to facial expressions, that can distinguish between identical twins (the first two authors). We demonstrate a prototype system based on the proposed algorithm and compare its performance to classical face recognition methods.The numerical methods employed by our approach do not require the facial surface explicitly. The surface gradients field, or the surface metric, are sufficient for constructing the expression-invariant representation of any given face. It allows us to perform the 3D face recognition task while avoiding the surface reconstruction stage.     

Face recognition using laplacianfaces

We propose an appearance-based face recognition method called the Laplacianface approach. By using locality preserving projections (LPP), the face images are mapped into a face subspace for analysis. Different from principal component analysis (PCA) and linear discriminant analysis (LDA) which effectively see only the Euclidean structure of face space, LPP finds an embedding that preserves local information, and obtains a face subspace that best detects the essential face manifold structure. The Laplacianfaces are the optimal linear approximations to the eigenfunctions of the Laplace Beltrami operator on the face manifold. In this way, the unwanted variations resulting from changes in lighting, facial expression, and pose may be eliminated or reduced. Theoretical analysis shows that PCA, LDA, and LPP can be obtained from different graph models. We compare the proposed Laplacianface approach with Eigenface and Fisherface methods on three different face data sets. Experimental results suggest that the proposed Laplacianface approach provides a better representation and achieves lower error rates in face recognition.     

Real-time recovery of moving 3D faces for emerging applications

Existing face imaging systems are not suitable to meet the face representation and recognition demands for emerging applications in areas such as interactive gaming, enhanced learning environments and directed advertising. This is mainly due to the poor capture and characterisation of facial data that compromises their spatial and temporal precision. For emerging applications it is not only necessary to have a high level of precision for the representation of facial data, but also to characterise dynamic faces as naturally as possible and in a timely manner. This study proposes a new framework for capturing and recovering dynamic facial information in real-time at significantly high order of spatial and temporal accuracy to capture and model subtle facial changes for enhanced realism in 3D face visualisation and higher precision for face recognition applications. We also present a novel, fast, and robust correspondence mapping approach for 3D registration of moving 3D faces.     

Face recognition in hyperspectral images

Hyperspectral cameras provide useful discriminants for human face recognition that cannot be obtained by other imaging methods. We examine the utility of using near-infrared hyperspectral images for the recognition of faces over a database of 200 subjects. The hyperspectral images were collected using a CCD camera equipped with a liquid crystal tunable filter to provide 31 bands over the near-infrared (0.7 /spl mu/m-1.0 /spl mu/m). Spectral measurements over the near-infrared allow the sensing of subsurface tissue structure which is significantly different from person to person, but relatively stable over time. The local spectral properties of human tissue are nearly invariant to face orientation and expression which allows hyperspectral discriminants to be used for recognition over a large range of poses and expressions. We describe a face recognition algorithm that exploits spectral measurements for multiple facial tissue types. We demonstrate experimentally that this algorithm can be used to recognize faces over time in the presence of changes in facial pose and expression.     

A novel application of self-organizing network for facial expression recognition from radial encoded contours

We propose an efficient algorithm for recognizing facial expressions using biologically plausible features: contours of face and its components with radial encoding strategy. A self-organizing network (SON) is applied to check the homogeneity of the encoded contours and then different classifiers, such as SON, multi-layer perceptron and K-nearest neighbor, are used for recognizing expressions from contours. Experimental results show that the recognition accuracy of our algorithm is comparable to that of other algorithms in the literature on the Japanese female facial expression database. We also apply our algorithm to Taiwanese facial expression image database to demonstrate its efficiency in recognizing facial expressions.     

Robust parametric active contours: the Sandwich Snakes

Snakes are active contours that minimize an energy function. We present a new kind of active contours called “Sandwich Snakes”. They are formed by two snakes, one inside and the other outside of the curve that one is looking for. They have the same number of particles, which are connected in one-to-one correspondence. At the minimum the two snakes have the same position. We also present here a multi-scale system, where Sandwich Snakes are adjusted at increasing resolutions, and an interactive tool that permits one to easily specify the initial position for the Sandwich Snakes. Sandwich Snakes exhibit very good perfomance detecting contours with complex shapes, where the traditional methods fail. They are also very robust with respect to noise. Received: 29 January 1999 / Accepted: 20 August 2000     

基于非负张量分解的人脸识别算法研究

人脸识别是生物特征识别中一个活跃的研究领域。非负张量分解作为非负矩阵分解的多线性推广,已被成功应用到人脸识别等领域。提出了基于非负张量分解的人脸识别算法。该方法无需将人脸矩阵向量化,从而保持了人脸矩阵的内部结构,即人脸图像的整体结构,使人脸特征提取更精确。 实验结果表明, 与经典的人脸识别算法如PCA和NMF相比,该算法提供了一种更好的脸部表示模式,提高了人脸识别的正确率。     

Open-set face recognition across look-alike faces in real-world scenarios

The open-set problem is among the problems that have significantly changed the performance of face recognition algorithms in real-world scenarios. Open-set operates under the supposition that not all the probes have a pair in the gallery. Most face recognition systems in real-world scenarios focus on handling pose, expression and illumination problems on face recognition. In addition to these challenges, when the number of subjects is increased for face recognition, these problems are intensified by look-alike faces for which there are two subjects with lower intra-class variations. In such challenges, the inter-class similarity is higher than the intra-class variation for these two subjects. In fact, these look-alike faces can be created as intrinsic, situation-based and also by facial plastic surgery. This work introduces three real-world open-set face recognition methods across facial plastic surgery changes and a look-alike face by 3D face reconstruction and sparse representation. Since some real-world databases for face recognition do not have multiple images per person in the gallery, with just one image per subject in the gallery, this paper proposes a novel idea to overcome this challenge by 3D modeling from gallery images and synthesizing them for generating several images. Accordingly, a 3D model is initially reconstructed from frontal face images in a real-world gallery. Then, each 3D reconstructed face in the gallery is synthesized to several possible views and a sparse dictionary is generated based on the synthesized face image for each person. Also, a likeness dictionary is defined and its optimization problem is solved by the proposed method. Finally, the face recognition is performed for open-set face recognition using three proposed representation classifications. Promising results are achieved for face recognition across plastic surgery and look-alike faces on three databases including the plastic surgery face, look-alike face and LFW databases compared to several state-of-the-art methods. Also, several real-world and open-set scenarios are performed to evaluate the proposed method on these databases in real-world scenarios.     

Application of semantic features in face recognition

We propose a new face recognition strategy, which integrates the extraction of semantic features from faces with tensor subspace analysis. The semantic features consist of the eyes and mouth, plus the region outlined by the centers of the three components. A new objective function is generated to fuse the semantic and tensor models for finding similarity between a face and its counterpart in the database. Furthermore, singular value decomposition is used to solve the eigenvector problem in the tensor subspace analysis and to project the geometrical properties to the face manifold. Experimental results demonstrate that the proposed semantic feature-based face recognition algorithm has favorable performance with more accurate convergence and less computational efforts.