Face Registration Research Articles

Frenet Frame-Based Generalized Space Curve Representation for Pose-Invariant Classification and Recognition of 3-D Face

The state-of-the-art methods in classifying 3-D representation of the face involve challenges in extracting representative features directly from the large volume of facial data. These methods mostly ignore the effect of pose distortions on 3-D facial data and entail heavy computations as well as manual processing steps. This work proposes a novel Frenet frame-based generalized space curve representation method for 3-D pose-invariant face and facial expression recognition and classification. Three-dimensional facial curves are extracted from either frontal or synthetically posed 3-D facial data to derive the proposed Frenet frame-based features. A mathematical framework shows the proof of pose invariance property for the features. The effectiveness of the proposed method is evaluated in two recognition tasks: 3-D face recognition (3D-FR) and 3-D facial expression recognition (3D-FER) using benchmarked 3-D datasets. The proposed framework yields 96% rank-I recognition rate for 3D-FR and 91.4% area under ROC curves for six basic 3D-FER. The performance evaluation also shows that the proposed mathematical framework yields pose-invariant 3D-FR and 3D-FER for a wide range of pose angles. This pose invariance property of the Frenet frame-based features alleviates the need for an expensive 3-D face registration in the preprocessing step, which, in turn, enables a faster processing time. The evaluation results further suggest that the proposed method is not only computationally efficient and versatile, but also offers competitive performance when compared with the existing state-of-the-art methods reported for either 3D-FR or 3D-FER.

Assessment of an image-guided neurosurgery system using a head phantom

Objective: To acknowledge the challenges and limitations of image-guided neurosurgery systems, we compared the application accuracy of two different image registration methods for one commercial system. (VectorVision, BrainLab, Germany).Methods: We used an anthropomorphic head phantom for radiosurgery and a custom built add-on to simulate surgical targets inside the brain during an image-guided neurosurgery. We used two image registration methods, fiducial registration using attachable surface markers for computed tomography (CT) and surface registration using infrared laser face scanning. After simulation, we calculated the three-dimensional (3D) distance between the predicted position of a target, and its actual position using a registered pointer and an infrared camera. Deviations were measured for both superficial fiducial markers and internal surgical targets by five different users.Results: Deviations from the location of fiducial markers after each registration method were 2.15 ± 0.93 mm after CT surface marker registration and 1.25 ± 0.64 mm after infrared face scanner registration. The mean target registration errors were 2.95 ± 1.4 mm using fiducial registration and 2.90 ± 1.3 mm using surface registration. The largest deviations (6.2 mm) were found for the targets in the skull base and posterior cranial fossa. Fiducial deviations and target registration errors were statistically uncorrelated. The total application accuracy was 4.87 ± 0.97 mm after CT surface marker registration and 4.14 ± 0.64 mm after infrared face scanner registration.Conclusions: Despite others have reported differences, we did not find significant variations between both registration methods for the target registration error, although application accuracy was slightly better after surface face registration. Superficial registration errors, but not the target registration error, can be routinely evaluated in the operating room. Since both errors were uncorrelated, surgeons may neglect the achievable accuracy of the procedure. The described method is recommended to assess application accuracy in the operating room.

Three-dimensional (3D) facial recognition and prediction

This paper provides solution to the problem in identifying humans from their 3D facial characteristics. For this reason, a standard 3D facial recognition system was built and used in this research work. This work proposes two novel fusion schemes where the first one employs a confidence-aided combination approach, and the second one implements a two-level serial integration method. Recognition simulations performed on the 3DRMA and the FRGC databases show that: (1) generic face template-based rigid registration of faces is better than the non-rigid variant, (2) principal curvature directions and surface normal have better discriminative power, (3) representing faces using local patch descriptors can both reduce the feature dimensionality and improve the identification rate, and (4) confidence-assisted fusion rules and serial two-stage fusion schemes have a potential to improve the accuracy when compared to other decision-level fusion rules.

Register-My-Face: a tool to register three-dimensional face images

Efficient registration across pose is the most challenging research area for accurate recognition of human face images. Authors have discussed a tool that has been developed for registration of face images across poses using the nose tip of the face images. The nose tip has been considered here because of its stability in situations such as variations in pose, expression, etc. The aim of this investigation is to develop a face registration tool called “Register-My-Face” with a working methodology in all the three directions, namely yaw, pitch, and roll. This tool has been developed for three-dimensional (3-D) face registration, which is inspired by analyzing the “depth values” of face range images. The registration of the face is done using a geometrical technique which is based on computing the corresponding rotation in three orthogonal directions. The advantages of the designed tool are that it does not need any training phase for accurate detection of the nose tip, and this method can handle large pose variations, including 90 deg pose variations about the Y-axis in both the positive and negative directions. The method that has been followed to develop this tool is also independent of facial expression, occlusion and illumination variations. Moreover, it quickly detects the nose tip because it does not need to process the entire face surface, but only requires the isolated nose region. The tool has been integrated with three different databases; GavabDB, Bosphorus, and Frav3D, and the investigation highlights the robustness of the tool. Additionally, for exploring the performance of the tool, a SIMULINK model for hardware interface is also developed with a discrete solver and is tested on two different configuration setups and executed in two different execution modes with two simulation stop timings 10.0 and 1.0. This model can proceed according to the algorithm with a minimum of 6.640 s to register an unregistered raw 3-D face scan input image from the Frav3D database. Accuracies of the nose region of 98.87%, 94.44%, and 98.08% for the Frav3D, GavabDB, and Bosphorus databases, respectively, are observed. For nose tip detection, the success rates are 98.91% for the Frav3D database, 98.74% for the GavabDB database, and 96.03% for the Bosphorus database. Based on the success rate of nose tip detection, the registration process is implemented on three databases. Registration accuracy, computed between a neutral and the registered range face image for the Frav3D database is 87.5% for GavabDB and 89.87% for Bosphorus database, and the rate of success is 70.23%.

3D face recognition based on pose and expression invariant alignment

In this paper we present a novel pose and expression invariant approach for 3D face registration based on intrinsic coordinate system characterized by nose tip, horizontal nose plane and vertical symmetry plane of the face. It is observed that distance of nose tip from 3D scanner is reduced after pose correction which is presented as a quantifying heuristic for proposed registration scheme. In addition, motivated by the fact that a single classifier cannot be generally efficient against all face regions, a two tier ensemble classifier based 3D face recognition approach is presented which employs Principal Component Analysis (PCA) for feature extraction and Mahalanobis Cosine (MahCos) matching score for classification of facial regions with weighted Borda Count (WBC) based combination and a re-ranking stage. The performance of proposed approach is corroborated by extensive experiments performed on two databases: GavabDB and FRGC v2.0, confirming effectiveness of fusion strategies to improve performance.

Automatic Analysis of Facial Affect: A Survey of Registration, Representation, and Recognition.

Automatic affect analysis has attracted great interest in various contexts including the recognition of action units and basic or non-basic emotions. In spite of major efforts, there are several open questions on what the important cues to interpret facial expressions are and how to encode them. In this paper, we review the progress across a range of affect recognition applications to shed light on these fundamental questions. We analyse the state-of-the-art solutions by decomposing their pipelines into fundamental components, namely face registration, representation, dimensionality reduction and recognition. We discuss the role of these components and highlight the models and new trends that are followed in their design. Moreover, we provide a comprehensive analysis of facial representations by uncovering their advantages and limitations; we elaborate on the type of information they encode and discuss how they deal with the key challenges of illumination variations, registration errors, head-pose variations, occlusions, and identity bias. This survey allows us to identify open issues and to define future directions for designing real-world affect recognition systems.

A new algorithm for non-rigid shape matching with anisotropic-scaling transformation parameters

Three-dimensional (3D) shape matching finds a wide application in manufacturing industry, such as surface inspection, workpiece localization and reverse engineering. In this paper, a new optimization method is proposed for non-rigid shape matching modeling and solving. The point-tangent distance function is used to construct a nonlinear least-squares model for anisotropic non-rigid shape matching, from which 9 variables with respect to non-rigid matching parameters are computed simultaneously by solving a linear system. In order to strengthen the non-rigid matching robustness to potential outliers, the optimization model is improved by an iteratively reweighted method. The typical characteristic is to weaken the influence of outliers during iterations. Finally, experiments are carried out to evaluate and analyze the proposed method, including matching accuracy, efficiency and robustness. Shape matching is also an important task in computer vision (such as face registration and object recognition), and the proposed method can find its application in this field.

Automatic visible and infrared face registration based on silhouette matching and robust transformation estimation

Registration of multi-sensor data (particularly visible color sensors and infrared sensors) is a prerequisite for multimodal image analysis such as image fusion. In this paper, we proposed an automatic registration technique for visible and infrared face images based on silhouette matching and robust transformation estimation. The key idea is to represent a (visible or infrared) face image by its silhouette which is extracted from the image’s edge map and consists of a set of discrete points, and then align the two silhouette point sets by using their feature similarity and spatial geometrical information. More precisely, our algorithm first matches the silhouette point sets by their local shape features such as shape context, which creates a set of putative correspondences that may contaminated by outliers. Next, we estimate the accurate transformation from the putative correspondence set under a robust maximum likelihood framework combining with the EM algorithm, where the transformation between the image pair is modeled by a parametric model such as the rigid or affine transformation. The qualitative and quantitative comparisons on a publicly available database demonstrate that our method significantly outperforms other state-of-the-art visible/infrared face registration methods. As a result, our method will be beneficial for fusion-based face recognition.

Nose tip detection on three‐dimensional faces using pose‐invariant differential surface features

Three-dimensional (3D) facial data offer the potential to overcome the difficulties caused by the variation of head pose and illumination in 2D face recognition. In 3D face recognition, localisation of nose tip is essential to face normalisation, face registration and pose correction etc. Most of the existing methods of nose tip detection on 3D face deal mainly with frontal or near-frontal poses or are rotation sensitive. Many of them are training-based or model-based. In this study, a novel method of nose tip detection is proposed. Using pose-invariant differential surface features – high-order and low-order curvatures, it can detect nose tip on 3D faces under various poses automatically and accurately. Moreover, it does not require training and does not depend on any particular model. Experimental results on GavabDB verify the robustness and accuracy of the proposed method.

3D and 2D face recognition using integral projection curves based depth and intensity images

This paper presents an automatic face recognition system in the presence of illumination, expressions and pose variations based on depth and intensity information. At first, the registration of 3D faces is achieved using iterative closest point ICP. Nose tip point must be located using Maximum Intensity Method. This point usually has the largest depth value; however there is a problem with some unnecessary data such as: shoulders, hair, neck and parts of clothes; to cope with this issue, we propose the integral projection curves IPC-based facial area segmentation to extract the facial area. After that, the combined method principal component analysis PCA with enhanced Fisher model EFM is used to obtain the feature matrix vectors. Finally, the classification is performed using distance measurement and support vector machine SVM. The experiments are implemented on two face databases CASIA3D and GavabDB; our results show that the proposed method achieves a high recognition performance.

Non-rigid visible and infrared face registration via regularized Gaussian fields criterion

Registration of multi-sensor data (particularly visible color sensors and infrared sensors) is a prerequisite for multimodal image analysis such as image fusion. Typically, the relationships between image pairs are modeled by rigid or affine transformations. However, this cannot produce accurate alignments when the scenes are not planar, for example, face images. In this paper, we propose a regularized Gaussian fields criterion for non-rigid registration of visible and infrared face images. The key idea is to represent an image by its edge map and align the edge maps by a robust criterion with a non-rigid model. We model the transformation between images in a reproducing kernel Hilbert space and a sparse approximation is applied to the transformation to avoid high computational complexity. Moreover, a coarse-to-fine strategy by applying deterministic annealing is used to overcome local convergence problems. The qualitative and quantitative comparisons on two publicly available databases demonstrate that our method significantly outperforms the state-of-the-art method with an affine model. As a result, our method will be beneficial for fusion-based face recognition.

Efficient smile detection by Extreme Learning Machine

Smile detection is a specialized task in facial expression analysis with applications such as photo selection, user experience analysis, and patient monitoring. As one of the most important and informative expressions, smile conveys the underlying emotion status such as joy, happiness, and satisfaction. In this paper, an efficient smile detection approach is proposed based on Extreme Learning Machine (ELM). The faces are first detected and a holistic flow-based face registration is applied which does not need any manual labeling or key point detection. Then ELM is used to train the classifier. The proposed smile detector is tested with different feature descriptors on publicly available databases including real-world face images. The comparisons against benchmark classifiers including Support Vector Machine (SVM) and Linear Discriminant Analysis (LDA) suggest that the proposed ELM based smile detector in general performs better and is very efficient. Compared to state-of-the-art smile detector, the proposed method achieves competitive results without preprocessing and manual registration.

Face Recognition using Average Half Face Template

Face recognition is one of the most important technologies, which has been well-developed over the last two decades. Face recognition technology has reached a level of utmost importance as the security issues increase worldwide. Most of the previously proposed systems, based on half face images are computationally slow and require more storage. In the proposed model, an average half face image is used for recognition to reduce computational time and storage requirements. The Viola Jones method is used in conjunction with intensity-based registration for real time face detection and registration, before splitting the full face. Finally, Principal Component Analysis (PCA) is used to compress the multi-dimensional data space and recognition. Experimental results clearly elaborate that half face recognition produces much better results as compared to the full face recognition and other previously proposed half face recognition models.

Teichmuller Mapping (T-Map) and Its Applications to Landmark Matching Registration

Registration, which aims to find an optimal 1-1 correspondence between shapes, is an important process in different research areas. Landmark-based surface registration has been widely studied to obtain a mapping between shapes that matches important features. Obtaining a unique and bijective surface registration that matches features consistently is generally challenging, especially when a large number of landmark constraints are enforced. This motivates us to search for a unique landmark matching surface diffeomorphism, which minimizes the local geometric distortion. For this purpose, we propose a special class of diffeomorphisms called the Teichmüller mappings (T-Maps). Under suitable conditions on the landmark constraints, a unique T-Map between two surfaces can be obtained, which minimizes the maximal conformality distortion. The conformality distortion measures how far the mapping deviates from a conformal mapping, and hence it measures the local geometric distortion. In this paper, we propose an efficient iterative algorithm, called the quasi-conformal (QC) iteration, to compute the T-Map. The basic idea is to represent the set of diffeomorphisms using Beltrami coefficients (BCs) and look for an optimal BC associated to the desired T-Map. The associated diffeomorphism can be efficiently reconstructed from the optimal BC using the linear Beltrami solver (LBS). Using BCs to represent diffeomorphisms guarantees the diffeomorphic property of the registration, even with very large deformation. Using our proposed method, the T-Map can be accurately and efficiently computed. The obtained registration is guaranteed to be bijective. The proposed algorithm can also be extended to compute T-Map with soft landmark constraints. We applied the proposed algorithm to real applications, such as brain landmark matching registration, constrained texture mapping, and human face registration. Experimental results shows that our method is both effective and efficient in computing a nonoverlap landmark matching registration with the least amount of conformality distortion.

A VIRTUAL ANALYSIS ON VARIOUS TECHNIQUES USING ANN WITH DATA MINING

In this paper, Firstly we discussed on monitoring t he quality of video in network and proposing a tool called “VQMT” (Video Quality Measurement Tool) for automatic assessment of video quality and comparing it with MOS (mean opinion sc ore). Secondly; author had proposed a tool called “ReGIMviZ” for video dat a visualization and personalization system based on semantic classification also used fuzzy logic. And lastly we focus on “SOFAIT” ( SIFT and Optical flow affine image Transform) techn ique for face registration in video to improve action unit and its various algori thms. Here, in every system the common area is ANN architecture based on supervised learning algorithm.

Detecting Genetic Association of Common Human Facial Morphological Variation Using High Density 3D Image Registration

Human facial morphology is a combination of many complex traits. Little is known about the genetic basis of common facial morphological variation. Existing association studies have largely used simple landmark-distances as surrogates for the complex morphological phenotypes of the face. However, this can result in decreased statistical power and unclear inference of shape changes. In this study, we applied a new image registration approach that automatically identified the salient landmarks and aligned the sample faces using high density pixel points. Based on this high density registration, three different phenotype data schemes were used to test the association between the common facial morphological variation and 10 candidate SNPs, and their performances were compared. The first scheme used traditional landmark-distances; the second relied on the geometric analysis of 15 landmarks and the third used geometric analysis of a dense registration of ∼30,000 3D points. We found that the two geometric approaches were highly consistent in their detection of morphological changes. The geometric method using dense registration further demonstrated superiority in the fine inference of shape changes and 3D face modeling. Several candidate SNPs showed potential associations with different facial features. In particular, one SNP, a known risk factor of non-syndromic cleft lips/palates, rs642961 in the IRF6 gene, was validated to strongly predict normal lip shape variation in female Han Chinese. This study further demonstrated that dense face registration may substantially improve the detection and characterization of genetic association in common facial variation.

A Priori Laplacian with Hamming Distance: Advanced Dimension Reduction Technique

Images containing faces are essential to intelligent visionbased human computer interaction, and research efforts in face processing include face recognition, face tracking, pose estimation, and expression recognition. The rapidly expanding research in face processing is based on the premise that information about a user’s identity, state, and intent can be extracted from images and that computers can then react accordingly, e.g., by knowing person’s identity, person may be authenticated to utilize a particular service or not. A first step of any face processing system is registering the locations in images where faces are present. However, face registration for whole database is a challenging task because of variability in scale, location, orientation (up-right, rotated), and pose (frontal, profile). Facial expression, occlusion, and lighting conditions also change the overall appearance of face. The Image registration algorithm will register all these images present in the database. The face recognition algorithm which is insensitive to large variation in lighting direction and facial expression is to be implemented. Taking a pattern classification approach, each pixel in an image can be considered as a coordinate in a high-dimensional space. The advantage of this is that the images of a particular face, under varying illumination but fixed pose, lie in a 3D linear subspace of the high dimensional image space—if the face is a Lambertian surface. However, since faces are not truly Lambertian surfaces and do indeed produce self-shadowing images will deviate from this linear subspace. Rather than explicitly modeling this deviation, project the image into a subspace in such a manner which discounts those regions of the face with large deviation. This is achieved by using dimension reduction techniques like Principal component analysis (PCA), Linear Discriminant analysis (LDA), Laplacian faces and other modified approaches like A Priori Laplacian and A Priori Laplacian with Hamming Distance. General Terms Face Recognition, Image Processing, Pattern Recognition

A neural network-based point registration method for 3D rigid face image

Intelligent detection of human face image combined with the real-time video monitoring has been applied to improve the secure and protective possibility. The registration is an indispensible step before distinguishing the variation among the images. Neural network (NN) has a strong learning ability from a mass unstructured point cloud even containing noisy data. Neural network has been applied to register 3D reconstructed ear data and 3D surface of bunny and to achieve the better results. Motivated by this idea, NN-based registration method for 3D rigid face image is proposed. This paper presented the proof process of obtaining rotation matrix and translation vector according to the training process of neural network. Then the measure index of registration performance was provided. The elaborate experiments were conducted on the 3D USF face database (provided by the University of South Florida) to verify the effectiveness of neural network as a registration method. Next, two comparisons were performed, one group was NN-based and ICP-based registration methods and the other group was our proposed NN-based and other NN-based registration methods. The experimental results showed that neural network is a robust and potential registration method for rigid face image registration. Furthermore, our proposed NN-based registration method is extended easily to do 2D-to-3D registration and non-rigid face registration.

Automatic landmark point detection and tracking for human facial expressions

Facial landmarks are a set of salient points, usually located on the corners, tips or mid points of the facial components. Reliable facial landmarks and their associated detection and tracking algorithms can be widely used for representing the important visual features for face registration and expression recognition. In this paper we propose an efficient and robust method for facial landmark detection and tracking from video sequences. We select 26 landmark points on the facial region to facilitate the analysis of human facial expressions. They are detected in the first input frame by the scale invariant feature based detectors. Multiple Differential Evolution-Markov Chain (DE-MC) particle filters are applied for tracking these points through the video sequences. A kernel correlation analysis approach is proposed to find the detection likelihood by maximizing a similarity criterion between the target points and the candidate points. The detection likelihood is then integrated into the tracker’s observation likelihood. Sampling efficiency is improved and minimal amount of computation is achieved by using the intermediate results obtained in particle allocations. Three public databases are used for experiments and the results demonstrate the effectiveness of our method.

Robust binary neural networks based 3D Face detection and accurate face registration

Abstract In this paper, we propose a facial feature localization algorithm based on a binary neural network technique - k-Nearest Neighbour Advanced Uncertain Reasoning Architecture (kNN AURA) to encode, train and match the feature patterns to accurate identify the nose tip in 3D. Based on the results of the 3D nose tip localization, the main face area is detected and cropped from the original 3D image. Then we present a novel framework to implement the 3D face registration by several integrated phases. First we use Principal Component Analysis (PCA) to roughly correct the server misalignment. Then we exploit the symmetric of human face to reduce the misalignment about oy and oz axis. In order to reduce the effect of facial expression variations, the expression-invariant region is segmented. Using Iterative Closest Point (ICP) algorithms, the expression-invariant region of faces can be aligned according to a standard face model, the misalignment about ox is then eventually corrected. Our experiments perfo...