Kwontaeg Choi

Realtime training on mobile devices for face recognition applications

Due to the increases in processing power and storage capacity of mobile devices over the years, an incorporation of realtime face recognition to mobile devices is no longer unattainable. However, the possibility of the realtime learning of a large number of samples within mobile devices must be established. In this paper, we attempt to establish this possibility by presenting a realtime training algorithm in mobile devices for face recognition related applications. This is differentiated from those traditional algorithms which focused on realtime classification. In order to solve the challenging realtime issue in mobile devices, we extract local face features using some local random bases and then a sequential neural network is trained incrementally with these features. We demonstrate the effectiveness of the proposed algorithm and the feasibility of its application in mobile devices through empirical experiments. Our results show that the proposed algorithm significantly outperforms several popular face recognition methods with a dramatic reduction in computational speed. Moreover, only the proposed method shows the ability to train additional samples incrementally in realtime without memory failure and accuracy degradation using a recent mobile phone model.

Incremental face recognition for large-scale social network services

Due to the rapid growth of social network services such as Facebook and Twitter, incorporation of face recognition in these large-scale web services is attracting much attention in both academia and industry. The major problem in such applications is to deal efficiently with the growing number of samples as well as local appearance variations caused by diverse environments for the millions of users over time. In this paper, we focus on developing an incremental face recognition method for Twitter application. Particularly, a data-independent feature extraction method is proposed via binarization of a Gabor filter. Subsequently, the dimension of our Gabor representation is reduced considering various orientations at different grid positions. Finally, an incremental neural network is applied to learn the reduced Gabor features. We apply our method to a novel application which notifies new photograph uploading to related users without having their ID being identified. Our extensive experiments show that the proposed algorithm significantly outperforms several incremental face recognition methods with a dramatic reduction in computational speed. This shows the suitability of the proposed method for a large-scale web service with millions of users.

Object tracking based on an online learning network with total error rate minimization

Abstract This paper presents a visual object tracking system which is tolerant to external imaging factors such as illumination, scale, rotation, occlusion and background changes. Specifically, an integration of an online version of total-error-rate minimization based projection network with an observation model of particle filter is proposed to effectively distinguish between the target object and the background. A re-weighting technique is proposed to stabilize the sampling of particle filter for stochastic propagation. For self-adaptation, an automatic updating scheme and extraction of training samples are proposed to adjust to system changes online. Our qualitative and quantitative experiments on 16 public video sequences show convincing performances in terms of tracking accuracy and computational efficiency over competing state-of-the-art algorithms.

A collaborative face recognition framework on a social network platform

Face recognition has many useful applications spanning surveillance, law enforcement, information security, smart card and entertainment technologies. Very recently, a learning based face recognition system is also seen to be applied to Web platform combining face recognition and Web service. However, many existing methods which focused on recognition accuracy cannot cope with the new social network platform because the adopted static learning approach is not adaptive to daily updated photographs among the massive number of users. In this paper, we discuss the difference between a stand-alone based system and a social network based system and propose a new collaborative face recognition framework where a redundant tagging can be avoided via sharing the identification information for efficient update under the social network platform. Our Experiments (including a Web stress test) using a public database show that the proposed method records a better accuracy than that of the state-of-the-art classifier SVM adopting a polynomial kernel and has fast execution time for both training and testing.

Extraction and fusion of partial face features for cancelable identity verification

In this paper, we propose to extract localized random features directly from partial face image matrix for cancelable identity verification. Essentially, the extracted random features consist of compressed horizontal and vertical facial information obtained from a structured projection of the raw face images. For template security reason, the face appearance information is concealed via averaging several templates over different transformations. The match score outputs of these cancelable templates are then fused through a total error rate minimization. Extensive experiments were carried out to evaluate and benchmark the performance of the proposed method based on the AR, FERET, ORL, Sheffield and BERC databases. Our empirical results show encouraging performances in terms of verification accuracy as well as satisfying four cancelable biometric properties.

Service-oriented architecture based on biometric using random features and incremental neural networks

We propose a service-oriented architecture based on biometric system where training and classification tasks are used by millions of users via internet connection. Such a large-scale biometric system needs to consider template protection, accuracy and efficiency issues. This is a challenging problem since there are tradeoffs among these three issues. In order to simultaneously handle these issues, we extract both global and local features via controlling the sparsity of random bases without training. Subsequently, the extracted features are fused with a sequential classifier. In the proposed system, the random basis features are not stored for security reason. The non-training based on feature extraction followed by a sequential learning contributes to computational efficiency. The overall accuracy is consequently improved via an ensemble of classifiers. We evaluate the performance of the proposed system using equal error rate under a stolen-token scenario. Our experimental results show that the proposed method is robust over severe local deformation with efficient computation for simultaneous transactions. Although we focus on face biometrics in this paper, the proposed method is generic and can be applied to other biometric traits.

A Random Network Ensemble for Face Recognition

In this paper, we propose a random network ensemble for face recognition problem, particularly for images with a large appearance variation and with a limited number of training set. In order to reduce the correlation within the network ensemble using a single type of feature extractor and classifier, localized random facial features have been constructed together with internally randomized networks. The ensemble classifier is finally constructed by combining these multiple networks via a sum rule. The proposed method is shown to have a better accuracy(31.5% and 15.3% improvements on AR and EYALEB databases respectively) and a better efficiency than that of the widely used PCA-SVM.

Face Alignment Using Segmentation and a Combined AAM in a PTZ Camera

In this paper, we propose a novel framework for face alignment based on the active appearance model (AAM) in surveillance systems with pan-tilt-zoom (PTZ) cameras. The AAM converges poorly in face images which are affected by illumination factors, cluttered backgrounds and status of the camera. To search for robust face model parameters, we propose a robust AAM fitting method based on segmenting faces and combining person-specific and generic models to achieve accurate face alignment. We segment faces using histogram back-projection and a skin color histogram, which is updated using a skin mask extracted by the AAM. For robust face recognition, we combined generic and person-specific models with a slight reduction in processing time. The extracted AAM parameters are as accurate as those when using the person-specific model and can be used as features for face recognition. Empirical experiments show that our proposed method extracts very accurate face parameters and is not sensitive to initial shapes

Visual tracking with online discriminative learning

We treat tracking as a binary classification task in order to distinguish between an object to be tracked and the background. We propose to integrate an online learning based total-error-rate minimization method (OTER) with an observation model of particle filter for visual tracking. The particle filter is modeled using an affine dynamic model and an observation model. The observation model is constructed using the OTER classifier for binary pattern classification. The proposed method is empirically evaluated both qualitatively and quantitatively using several publicly available video sequences.

Coarse-to-Fine Particle Filter by Implicit Motion Estimation for 3D Head Tracking on Mobile Devices

Due to the widely spread mobile devices over the years, a low cost implementation of an efficient head tracking system is becoming more useful for a wide range of applications. In this paper, we make an attempt to solving real-time 3D head tracking problem on mobile devices by enhancing the fitness of the dynamics. In our method, the particles are generated by implicit motion estimation between two particles rather than the explicit motion estimation using corresponding point matching between consecutive two frames. This generation is applied iteratively using coarse-to fine strategy in order to handle a large motion using a small number of particle. This reduces the computational cost while preserving the performance. We evaluate the efficiency and effectiveness of the proposed algorithm by empirical experiments. Finally, we demonstrate our method on a recent mobile phone.