Masato Kawade

Tracking in Low Frame Rate Video: A Cascade Particle Filter with Discriminative Observers of Different Life Spans

Tracking object in low frame rate video or with abrupt motion poses two main difficulties which most conventional tracking methods can hardly handle: 1) poor motion continuity and increased search space; 2) fast appearance variation of target and more background clutter due to increased search space. In this paper, we address the problem from a view which integrates conventional tracking and detection, and present a temporal probabilistic combination of discriminative observers of different lifespans. Each observer is learned from different ranges of samples, with different subsets of features, to achieve varying level of discriminative power at varying cost. An efficient fusion and temporal inference is then done by a cascade particle filter which consists of multiple stages of importance sampling. Experiments show significantly improved accuracy of the proposed approach in comparison with existing tracking methods, under the condition of low frame rate data and abrupt motion of both target and camera.

Person-Specific SIFT Features for Face Recognition

Scale invariant feature transform (SIFT) proposed by Lowe has been widely and successfully applied to object detection and recognition. However, the representation ability of SIFT features in face recognition has rarely been investigated systematically. In this paper, we proposed to use the person-specific SIFT features and a simple non-statistical matching strategy combined with local and global similarity on key-points clusters to solve face recognition problems. Large scale experiments on FERET and CAS-PEAL face databases using only one training sample per person have been carried out to compare it with other non person-specific features such as Gabor wavelet feature and local binary pattern feature. The experimental results demonstrate the robustness of SIFT features to expression, accessory and pose variations.

Tracking in Low Frame Rate Video: A Cascade Particle Filter with Discriminative Observers of Different Lifespans

Tracking object in low frame rate video or with abrupt motion poses two main difficulties which conventional tracking methods can barely handle: 1) poor motion continuity and increased search space; 2) fast appearance variation of target and more background clutter due to increased search space. In this paper, we address the problem from a view which integrates conventional tracking and detection, and present a temporal probabilistic combination of discriminative observers of different lifespans. Each observer is learned from different ranges of samples, with different subsets of features, to achieve varying level of discriminative power at varying cost. An efficient fusion and temporal inference is then done by a cascade particle filter which consists of multiple stages of importance sampling. Experiments show significantly improved accuracy of the proposed approach in comparison with existing tracking methods, under the condition of low frame rate data and abrupt motion of both target and camera.

Ethnicity estimation with facial images

We have advanced an effort to develop vision based human understanding technologies for realizing human-friendly machine interfaces. Visual information, such as gender, age ethnicity, and facial expression play an important role in face-to-face communication. This paper addresses a novel approach for ethnicity classification with facial images. In this approach, the Gabor wavelets transformation and retina sampling are combined to extract key facial features, and support vector machines that are used for ethnicity classification. Our system, based on this approach, has achieved approximately 94% for ethnicity estimation under various lighting conditions.

Discriminant analysis in correlation similarity measure space

Correlation is one of the most widely used similarity measures in machine learning like Euclidean and Mahalanobis distances. However, compared with proposed numerous discriminant learning algorithms in distance metric space, only a very little work has been conducted on this topic using correlation similarity measure. In this paper, we propose a novel discriminant learning algorithm in correlation measure space, Correlation Discriminant Analysis (CDA). In this framework, based on the definitions of within-class correlation and between-class correlation, the optimum transformation can be sought for to maximize the difference between them, which is in accordance with good classification performance empirically. Under different cases of the transformation, different implementations of the algorithm are given. Extensive empirical evaluations of CDA demonstrate its advantage over alternative methods.

Incremental Learning of Boosted Face Detector

In recent years, boosting has been successfully applied to many practical problems in pattern recognition and computer vision fields such as object detection and tracking. As boosting is an offline training process with beforehand collected data, once learned, it cannot make use of any newly arriving ones. However, an offline boosted detector is to be exploited online and inevitably there must be some special cases that are not covered by those beforehand collected training data. As a result, the inadaptable detector often performs badly in diverse and changeful environments which are ordinary for many real-life applications. To alleviate this problem, this paper proposes an incremental learning algorithm to effectively adjust a boosted strong classifier with domain-partitioning weak hypotheses to online samples, which adopts a novel approach to efficient estimation of training losses received from offline samples. By this means, the offline learned general-purpose detectors can be adapted to special online situations at a low extra cost, and still retains good generalization ability for common environments. The experiments show convincing results of our incremental learning approach on challenging face detection problems with partial occlusions and extreme illuminations.

Sparse Bayesian Regression for Head Pose Estimation

This paper presents a high performance head pose estimation system based on the newly-proposed sparse Bayesian regression technique (relevance vector machine, RVM) and sparse representation of facial patterns. In our system, after localizing 20 key facial points, sparse features of these points are extracted to represent facial property, and then RVM is utilized to learn the relation between the sparse representation and yaw and pitch angle. Because RVM requires only a very few kernel functions, it can guarantee better generalization, faster speed and less memory in a practical implementation. To thoroughly evaluate the performance of our system, we compare it with conventional methods such as CCA, kernel CCA, SVR on a large database; In experiments, we also investigate the influence of the facial points localization error on pose estimation by using manually labelled results and automatically localized results separately, and the influence of different features on pose estimation such as geometrical features and texture features. These experimental results demonstrate that our system can estimate face pose more accurately, robustly and fast than those based on conventional methods

Robust facial feature point detection under nonlinear illuminations

Addresses the problem of facial feature point detection under different lighting conditions. Our goal is to develop an efficient detection algorithm, which is suitable for practical applications. The problems that we need to overcome include (1) high detection accuracy, (2) low computational time and (3) nonlinear illumination. An algorithm is developed and reported in the paper. One of the key factors affecting the performance of feature point detection is the accuracy in locating face boundary. To solve this problem, we propose to make use of skin color, lip color and also the face boundary information. The basic idea to overcome the nonlinear illumination is that, each person shares the same/similar facial primitives, such as two eyes, one nose and one mouth. So the binary images of each person should be similar. Again, if a binary image (with appropriate thresholding) is obtained from the gray scale image, the facial feature points can also be detection easily. To achieve this, we propose to use the integral optical density (IOD) on face region. We propose to use the average IOD to detect feature windows. As all the above-mentioned techniques are simple and efficient, the proposed method is computationally effective and suitable for practical applications. 743 images from the Omron database with different facial expressions, different glasses and different hairstyle captured indoor and outdoor have been used to evaluate the proposed method and the detection accuracy is around 86%. The computational time in Pentium III 750 MHz using matlab for implementation is less than 7 seconds.

Online Real Boosting for Object Tracking Under Severe Appearance Changes and Occlusion

Robust visual tracking is always a challenging but yet intriguing problem owing to the appearance variability of target objects. In this paper we propose a novel method to handle large changes in appearance based on online real-value boosting, which is utilized to incrementally learn a strong classifier to distinguish between objects and their background. By incorporating online real boosting into a particle filter framework, our tracking algorithm shows a strong adaptability for different target objects which undergo severe appearance changes during the tracking process.

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.