Face recognition based on 3D ridge images obtained from range data

In this paper, we present an approach for 3D face recognition from frontal range data based on the ridge lines on the surface of the face. We use the principal curvature, k_max, to represent the face image as a 3D binary image called ridge image. The ridge image shows the locations of the ridge points around the important facial regions on the face (i.e., the eyes, the nose, and the mouth). We utilized the robust Hausdorff distance and the iterative closest points (ICP) for matching the ridge image of a given probe image to the ridge images of the facial images in the gallery. To evaluate the performance of our approach for 3D face recognition, we performed experiments on GavabDB face database (a small size database) and Face Recognition Grand Challenge V2.0 (a large size database). The results of the experiments show that the ridge lines have great capability for 3D face recognition. In addition, we found that as long as the size of the database is small, the performance of the ICP-based matching and the robust Hausdorff matching are comparable. But, when the size of the database increases, ICP-based matching outperforms the robust Hausdorff matching technique.     

Pose Robust Low-resolution Face Recognition via Coupled Kernel-based Enhanced Discriminant Analysis

Most face recognition techniques have been successful in dealing with high-resolution (HR) frontal face images. However, real-world face recognition systems are often confronted with the low-resolution (LR) face images with pose and illumination variations. This is a very challenging issue, especially under the constraint of using only a single gallery image per person. To address the problem, we propose a novel approach called coupled kernel-based enhanced discriminant analysis (CKEDA). CKEDA aims to simultaneously project the features from LR non-frontal probe images and HR frontal gallery ones into a common space where discrimination property is maximized. There are four advantages of the proposed approach: 1) by using the appropriate kernel function, the data becomes linearly separable, which is beneficial for recognition; 2) inspired by linear discriminant analysis (LDA), we integrate multiple discriminant factors into our objective function to enhance the discrimination property; 3) we use the gallery extended trick to improve the recognition performance for a single gallery image per person problem; 4) our approach can address the problem of matching LR non-frontal probe images with HR frontal gallery images, which is difficult for most existing face recognition techniques. Experimental evaluation on the multi-PIE dataset signifies highly competitive performance of our algorithm.      

可变光照条件下的人脸图像识别

对于人脸图像识别中光照变化的影响,传统的解决方法是对待识别图像进行光照补偿,先使它成为标准光照条件下的图像,然后和模板图像匹配来进行识别。为了提高在光照条件大范围变化时,人脸图像的识别率,提出了一种新的可变光照条件下的人脸图像识别方法。该方法首先利用在9个基本光照方向下分别获得的9幅图像来构成人脸光照特征空间,再通过这个光照特征空间,将图像库中的人脸图像变换成与待识别图像具有相同光照条件的图像,并将其作为模板图像;然后利用特征脸方法进行识别。实验结果表明,这种方法不仅能够有效地解决人脸识别中由于光照变化影响所造成的识别率下降的问题,而且对于光照条件大范围变化的情况,也可以得到比较高的正确识别率。     

On designing practical long range near infrared-based face recognition systems

Although automated face recognition (AFR) is a well-studied problem with a history of more than three decades, it is still far from being considered a solved problem for the case of difficult exposure conditions, such as during night-time, in environments with unconstrained lighting, or at large distances from the camera. However, in practical forensic scenarios, it is often the case that investigators operate in difficult conditions, where cross-session data need to be matched and where, grouping of the data in the context of demographic information (constitute the grouping in terms of gender, ethnicity) may be used in order to assist law enforcement officials, forensic investigators and security personnel in human identification practices. In this paper, we discuss the challenges in designing a practical near infrared (NIR) FR system and, more specifically, study the problems of intra-spectral, cross-spectral, i.e. VIS–NIR, intra-distance and cross-distance NIR FR, in indoors, outdoors, day-time and night-time environments. Furthermore, we propose the usage of a multi-feature scenario dependent fusion scheme that can enhance recognition performance. We also investigate which scenarios used, related to datasets, features useful for face matching or their combination, are most beneficial to the identification accuracy of NIR FR systems, when the gallery set is composed of either visible or NIR band face images. Thus, we illustrate that the selection of specific feature extraction techniques and their fusion are often the key design aspects that can turn practically non-functional systems to effective systems with real-world applicability. As a result, such a strategy can significantly extend the range of conditions under which automated NIR FR systems can operate.     

基于面部曲线弹性匹配的三维人脸识别方法

提出一种基于面部径向曲线弹性匹配的三维人脸识别方法。使用人脸曲 面上的多条曲线表征人脸曲面,提取三维人脸上从鼻尖点发射的多条面部径向曲线,对其进 行分层弹性匹配和点距对应匹配,根据人脸不同部位受表情影响程度不同,对不同曲线识别 相似度赋予不同权重进行加权融合作为总相似度用于识别。测试结果表明该方法具有很好的 识别性能,并且对表情、遮挡和噪声具有较好的鲁棒性。     

Human eyebrow recognition in the matching-recognizing framework

This paper studies the problem of automatically recognizing human eyebrows using a frontal view. In the matching-recognizing framework for image-based object classification, we design an automatic human eyebrow recognition system via fast template matching and Fourier spectrum distance. Fast template matching is used to locate the target subregion of a gallery template or a pure eyebrow image in a probe original eyebrow image, whereas Fourier spectrum distance is used to determine the final identity of the probe original eyebrow image. We conducted a number of experiments to demonstrate the efficacy of the proposed system and corroborate the validity of eyebrow recognition on the BJUT eyebrow database. Moreover, we also tested the system on the color FERET database. Experimental results show that our approach can be directly applied to face recognition by only replacing eyebrow templates with face templates, and may achieve higher accuracy in eyebrow recognition than in small face recognition. This is a strong argument for eyebrow recognition to replace face recognition as an independent biometric in certain scenarios, especially where relatively large eyebrows can be cropped.     

基于脊谷特征提取的三维人脸识别

针对三维人脸数据庞大及识别效率低的问题,提出采用提取脊点及谷点表征人脸。脊点和谷点作为曲面局部区域内主曲率沿主方向变化的极值点,能够很好地表征三维人脸特征。对三维人脸提取脊点模型和谷点模型,通过对它们栅格化后生成对应的空间分布密度直方图实现人脸粗匹配,采用计算LTS-Hausdorff距离实现人脸的精确匹配。在GavabDB三维人脸库的实验结果表明,该方法具有较高的识别率。     

A fast algorithm for ICP-based 3D shape biometrics

In a biometrics scenario, gallery images are enrolled into the database ahead of the matching step, which gives us the opportunity to build related data structures before the probe shape is examined. In this paper, we present a novel approach, called “Pre-computed Voxel Nearest Neighbor”, to reduce the computational time for shape matching in a biometrics context. The approach shifts the heavy computation burden to the enrollment stage, which is done offline. Experiments in 3D ear biometrics with 369 subjects and 3D face biometrics with 219 subjects demonstrate the effectiveness of our approach.     

Virtual illumination grid for correction of uncontrolled illumination in facial images

Face recognition under uncontrolled illumination conditions is still considered an unsolved problem. In order to correct for these illumination conditions, we propose a virtual illumination grid (VIG) approach to model the unknown illumination conditions. Furthermore, we use coupled subspace models of both the facial surface and albedo to estimate the face shape. In order to obtain a representation of the face under frontal illumination, we relight the estimated face shape. We show that the frontal illuminated facial images achieve better performance in face recognition. We have performed the challenging Experiment 4 of the FRGCv2 database, which compares uncontrolled probe images to controlled gallery images. Our illumination correction method results in considerably better recognition rates for a number of well-known face recognition methods. By fusing our global illumination correction method with a local illumination correction method, further improvements are achieved.     

COMPARING FACE IMAGES USING THE MODIFIED HAUSDORFF DISTANCE

We introduce a novel methodology applicable to face matching and fast screening of large facial databases. The proposed shape comparison method operates on edge maps and derives holistic similarity measures, yet, it does not require solving the point correspondence problem. While the use of edge images is important to introduce robustness to changes in illumination, the lack of point-to-point matching delivers speed and tolerance to local non-rigid distortions. In particular, we propose a face similarity measure derived as a variant of the Hausdorff distance by introducing the notion of a neighborhood function (N) and associated penalties (P). Experimental results on a large set of face images demonstrate that our approach produces excellent recognition results even when less than 3% of the original grey-scale face image information is stored in the face database (gallery). These results implicate that the process of face recognition may start at a much earlier stage of visual processing than it was earlier suggested. We argue, that edge-like retinal images of faces are initially screened “at a glance” without the involvement of high-level cognitive functions thus delivering high speed and reducing computational complexity.