Kai Zhu

Bio: Kai Zhu is an academic researcher from The Chinese University of Hong Kong. The author has contributed to research in topics: Gaussian process & Motion compensation. The author has an hindex of 2, co-authored 3 publications receiving 20 citations.

Orthogonal Gaussian Process for Automatic Age Estimation

Abstract: Age Estimation from facial images has been receiving increasing interest due to its important applications. Among the existing age estimation algorithms, the personalized approaches have been shown to be the most effective ones. However, most of the person-specific approaches (e.g. MTWGP [1], AGES [2], WAS [3]) rely heavily on the availability of training images across different ages for a single subject, which is very difficult to satisfy in practical applications. In order to overcome this problem, we propose a new approach to age estimation, called Orthogonal Gaussian Process (OGP). Compared to standard Gaussian Process, OGP is much more efficient while maintaining the discriminatory power of the standard Gaussian Process. Based on OGP, we further propose an improvement of OGP called anisotropic OGP (A-OGP) to enhance the age estimation performance. Extensive experiments are conducted to demonstrate the state-of-the-art estimation accuracy of our new algorithm on several public-domain face aging datasets: FG-NET face dataset with 82 different subjects, Morph Album 1 dataset with more than 600 subjects, and Morph Album 2 with about 20,000 different subjects.

A semantic model for video based face recognition

Abstract: Video-based face recognition has attracted a great deal of attention in recent years due to its wide applications. The challenge of video-based face recognition comes from several aspects. First, video data involves many frames, which increases data size and processing complexity. Second, key frames extracted from videos are usually of high intra-personal discrepancy due to variations in expressions, poses, and illuminations. In order to address these problems, we propose a novel semantic based subspace model to improve the performance of video based face recognition. The basic idea is to construct an appropriate low-dimensional subspace for each person, upon which a semantic model is built to classify the key frames of the person into specific class. After the semantic classification, the key frames belonging to the same classes, i.e. the same semantics, are used to train the linear classifiers for recognition. Extensive experiments on a large face video database (XM2VTS) clearly show that our approach obtains a significant performance improvement over the traditional approaches.

Multifeature Anisotropic Orthogonal Gaussian Process for Automatic Age Estimation

Abstract: Automatic age estimation is an important yet challenging problem. It has many promising applications in social media. Of the existing age estimation algorithms, the personalized approaches are among the most popular ones. However, most person-specific approaches rely heavily on the availability of training images across different ages for a single subject, which is usually difficult to satisfy in practical application of age estimation. To address this limitation, we first propose a new model called Orthogonal Gaussian Process (OGP), which is not restricted by the number of training samples per person. In addition, without sacrifice of discriminative power, OGP is much more computationally efficient than the standard Gaussian Process. Based on OGP, we then develop an effective age estimation approach, namely anisotropic OGP (A-OGP), to further reduce the estimation error. A-OGP is based on an anisotropic noise level learning scheme that contributes to better age estimation performance. To finally optimize the performance of age estimation, we propose a multifeature A-OGP fusion framework that uses multiple features combined with a random sampling method in the feature space. Extensive experiments on several public domain face aging datasets (FG-NET, MORPH Album1, and MORPH Album 2) are conducted to demonstrate the state-of-the-art estimation accuracy of our new algorithms.

Face recognition based on image sets

Abstract: In this thesis, we study the problem of face recognition based on image sets. The main objective of our work is to develop set-based distance metrics that are able to measure the similarity between image sets, rather than conventional distance metrics that can only measure the distance between samples. The face images obtained from real-life impose great challenges to the conventional face recognition systems. Large variations in appearances and various imperfections such as occlusions and misalignments in the face images severely degrade the recognition performance. One possible solution is to utilize more face images for each person, e.g., a collection of photos from personal galleries or frames extracted from a video clip. Under such circumstances, the face recognition task becomes the process of modelling and matching image sets. Our investigation then focuses on developing appropriate models and set-based distance metrics for representing different image sets.

MORPH: A Longitudinal Image Database of Normal Adult Age-Progression.

Abstract: This paper details MORPH a longitudinal face database developed for researchers investigating all facets of adult age-progression, e.g. face modeling, photo-realistic animation, face recognition, etc. This database contributes to several active research areas, most notably face recognition, by providing: the largest set of publicly available longitudinal images; longitudinal spans from a few months to over twenty years; and, the inclusion of key physical parameters that affect aging appearance. The direct contribution of this data corpus for face recognition is highlighted in the evaluation of a standard face recognition algorithm, which illustrates the impact that age-progression, has on recognition rates. Assessment of the efficacy of this algorithm is evaluated against the variables of gender and racial origin. This work further concludes that the problem of age-progression on face recognition (FR) is not unique to the algorithm used in this work.

Aging Face Recognition: A Hierarchical Learning Model Based on Local Patterns Selection

Abstract: Aging face recognition refers to matching the same person’s faces across different ages, e.g., matching a person’s older face to his (or her) younger one, which has many important practical applications, such as finding missing children. The major challenge of this task is that facial appearance is subject to significant change during the aging process. In this paper, we propose to solve the problem with a hierarchical model based on two-level learning. At the first level, effective features are learned from low-level microstructures, based on our new feature descriptor called local pattern selection (LPS). The proposed LPS descriptor greedily selects low-level discriminant patterns in a way, such that intra-user dissimilarity is minimized. At the second level, higher level visual information is further refined based on the output from the first level. To evaluate the performance of our new method, we conduct extensive experiments on the MORPH data set (the largest face aging data set available in the public domain), which show a significant improvement in accuracy over the state-of-the-art methods.

Non-Linear Dimensionality Reduction and Gaussian Process Based Classification Method for Smoke Detection

Abstract: To improve smoke detection accuracy, we combine local binary pattern (LBP) like features, kernel principal component analysis (KPCA), and Gaussian process regression (GPR) to propose a novel data processing pipeline for smoke detection. The data processing pipeline consists of three steps including original feature extraction, dimensionality reduction, and classification. We use LBP-like methods to extract original features. To obtain a more discriminant feature, KPCA is used to non-linearly map the original features into a discriminant feature space, where manifold structures are embedded. Finally, in order to improve generalization performance, we apply GPR to model classification as a Gaussian process by imposing Gaussian priors on both data and hyper-parameters. In addition, we can replace any steps of the pipeline by similar methods for further improvement or exploration, so the pipeline is flexible and extensible. Experimental results show that KPCA and GPR are truly able to improve the performance of smoke detection and texture classification, and our method obviously outperforms the same features with Support Vector Machine (SVM).