Yoshihisa Ijiri

Security Management for Mobile Devices by Face Recognition

Nowadays mobile devices maintain a lot of private information such as payment information, personal photographs and so on. In Japan, people are starting to use cellular phones as means of payment like pre-paid cards. If owners lost these devices and someone else picked it up, what would happen? To solve this problem, password is now widely used. In the case of password, it has to be complicated and long enough from the viewpoint of security, however it is troublesome for users to type such a long password. If someone set very easy password to avoid the troublesome process, it would be of no use. To solve both security and usability problem simultaneously, various types of biometrics have been proposed. Among many biometric solutions, since nowadays almost all of cellular phones have cameras, we propose face recognition that utilize these cameras. Through the verification process, users have only to take their facial photo using a camera equipped on mobile devices and wait for about one second. The system is effective enough to be used in the mobile devices in terms of performance, usability and hardware requirement.

Domain-Partitioning Rankboost for Face Recognition

In this paper we propose a domain partitioning RankBoost approach for face recognition. This method uses Local Gabor Binary Pattern Histogram (LGBPH) features for face representation, and adopts RankBoost to select the most discriminative features. Unlike the original RankBoost algorithm in Freund et al. (2003), weak hypotheses in our method make their predictions based on a partitioning of the similarity domain. Since the domain partitioning approach handles the loss function of a ranking problem directly, it can achieve a higher convergence speed than the original approach. Furthermore, in order to improve the algorithms generalization ability, we introduce some constraints to the weak classifiers being searched. Experiment results on FERET database show the effectiveness of our approach.

Fast and precise template matching based on oriented gradients

In this paper we propose a fast template matching method which can handle various types of objects. In our method the discretized orientations of image gradients which are robust to illumination changes and clutterd backgrounds are used as features. The features are binary represented and they can be matched very fast using bitwise operations. Furthermore, the rotated and resized templates those have similar feature vectors are clustered to one template and the total number of templates are greatly reduced, which boosts the detection speed. The experimental results show that our method can detect target objects (the search space includes translation, ±180 deg rotation, and ±50% scale change) with sub-pixel accracy in real-time.

Efficient Facial Attribute Recognition with a Spatial Codebook

There is a large number of possible facial attributes such as hairstyle, with/without glasses, with/without mustache, etc. Considering large number of facial attributes and their combinations, it is difficult to build attributes classifiers for all possible combinations needed in various applications, especially at the designing stage. To tackle this important and challenging problem, we propose a novel efficient facial attributes recognition algorithm using a learned spatial codebook. The Maximum Entropy and Maximum Orthogonality (MEMO) criterion is followed to learn the spatial codebook. With a spatial codebook constructed at the designing stage, attribute classifiers can be trained on demand with a small number of exemplars with high accuracy on the testing data. Meanwhile, up to 600 times speedup is achieved in the on-demand training process, compared to current state-of-the-art method. The effectiveness of the proposed method is supported by convincing experimental results.

Textureless object detection using cumulative orientation feature

We propose a novel image feature for textureless object detection. The feature is based on quantized gradient orientations those have been shown to be robust to cluttered backgrounds and illumination changes. We make this feature robust to the appearance changes of a targeted object itself induced by its transformations and small deformations. In our proposed method, we add small random values to the similarity transformation parameters and synthesize many model images. Then quantized orientations are extracted on these images and the orientations are cumulated at each pixel. The frequencies of selected features are utilized as weights when calculating scores. Our proposed feature is evaluated on publicly available dataset and achieve top-class performance both in speed and detection accuracy compared to state-of-the-art techniques.

Multiple Non-rigid Surface Detection and Registration

Detecting and registering nonrigid surfaces are two important research problems for computer vision. Much work has been done with the assumption that there exists only one instance in the image. In this work, we propose an algorithm that detects and registers multiple nonrigid instances of given objects in a cluttered image. Specifically, after we use low level feature points to obtain the initial matches between templates and the input image, a novel high-order affinity graph is constructed to model the consistency of local topology. A hierarchical clustering approach is then used to locate the nonrigid surfaces. To remove the outliers in the cluster, we propose a deterministic annealing approach based on the Thin Plate Spline (TPS) model. The proposed method achieves high accuracy even when the number of outliers is nineteen times larger than the inliers. As the matches may appear sparsely in each instance, we propose a TPS based match growing approach to propagate the matches. Finally, an approach that fuses feature and appearance information is proposed to register each nonrigid surface. Extensive experiments and evaluations demonstrate that the proposed algorithm achieves promising results in detecting and registering multiple non-rigid surfaces in a cluttered scene.

Deformed and Touched Characters Recognition

In this demonstration, we will show our Optical Character Recognition(OCR) technique. Character deformation and touching problems often occur during high-speed printing process in the machine vision industry. As a result, it is difficult for OCR system to segment and recognize characters properly. To solve these problems, we propose a novel OCR technique which is robust against deformation and touching. It splits regions of characters simply and excessively, recognizes all segments and merged regions, and obtains optimal segments using graph theory.

An adaptive nonparametric discriminant analysis method and its application to face recognition

Linear Discriminant Analysis (LDA) is frequently used for dimension reduction and has been successfully utilized in many applications, especially face recognition. In classical LDA, however, the definition of the between-class scatter matrix can cause large overlaps between neighboring classes, because LDA assumes that all classes obey a Gaussian distribution with the same covariance. We therefore, propose an adaptive nonparametric discriminant analysis (ANDA) algorithm that maximizes the distance between neighboring samples belonging to different classes, thus improving the discriminating power of the samples near the classification borders. To evaluate its performance thoroughly, we have compared our ANDA algorithm with traditional PCA+LDA, Orthogonal LDA (OLDA) and nonparametric discriminant analysis (NDA) on the FERET and ORL face databases. Experimental results show that the proposed algorithm outperforms the others.

Re-weighting Linear Discrimination Analysis under ranking loss

Linear discrimination analysis (LDA) is one of the most popular feature extraction and classifier design techniques. It maximizes the Fisher-ratio between between-class scatter matrix and within-class scatter matrix under a linear transformation, and the transformation is composed of the generalized eigenvectors of them. However, Fisher criterion itself can not decide the optimum norm of transformation vectors for classification. In this paper, we show that actually the norm of the transformation vectors has strong influence on classification performance, and we propose a novel method to estimate the optimum norm of LDA under the ranking loss, re-weighting LDA. On artificial data and real databases, the experiments demonstrate the proposed method can effectively improve the performance of LDA classifiers. And the algorithm can also be applied to other LDA variants such as non parametric discriminant analysis (NDA) to improve theirs performance further.