Xiaolong Wang

Deeply-Learned Feature for Age Estimation

Human age provides key demographic information. It is also considered as an important soft biometric trait for human identification or search. Compared to other pattern recognition problems (e.g., object classification, scene categorization), age estimation is much more challenging since the difference between facial images with age variations can be more subtle and the process of aging varies greatly among different individuals. In this work, we investigate deep learning techniques for age estimation based on the convolutional neural network (CNN). A new framework for age feature extraction based on the deep learning model is built. Compared to previous models based on CNN, we use feature maps obtained in different layers for our estimation work instead of using the feature obtained at the top layer. Additionally, a manifold learning algorithm is incorporated in the proposed scheme and this improves the performance significantly. Furthermore, we also evaluate different classification and regression schemes in estimating age using the deep learned aging pattern (DLA). To the best of our knowledge, this is the first time that deep learning technique is introduced and applied to solve the age estimation problem. Experimental results on two datasets show that the proposed approach is significantly better than the state-of-the-art.

Kinship Measurement on Salient Facial Features

Humans have the capability to recognize family members. Phrases such as “John has his fathers nose” or “Joe has his mothers eyes” are quite common. Motivated by this, we consider the following question: Is it possible to develop a method to extract the salient familial traits in face images for kinship recognition? If this idea works, an instrument may be invented to measure familial relationships. This computational kinship measurement might have a large impact in real applications, such as child adoptions, trafficking/smuggling of children, and finding missing children. The novel problem is related to but very different from traditional face recognition. It is more challenging than a typical face recognition problem since we need to find subtle features that are reliable across a large span of ages (e.g., grandfather and grandson) and sex difference (e.g., mother and son). A recently developed descriptor, i.e., DAISY, is adapted to our problem to represent the salient features, and a dynamic scheme is developed to stochastically combine familial traits. Experiments are performed on a database to show that our new approach can perform reasonably well for kinship verification. The encouraging result may inspire further research on this emerging problem.

A study on human age estimation under facial expression changes

In this paper, we study human age estimation in face images under significant expression changes. We will address two issues: (1) Is age estimation affected by facial expression changes and how significant is the influence? (2) How to develop a robust method to perform age estimation undergoing various facial expression changes? This systematic study will not only discover the relation between age estimation and expression changes, but also contribute a robust solution to solve the problem of cross-expression age estimation. This study is an important step towards developing a practical and robust age estimation system that allows users to present their faces naturally (with various expressions) rather than constrained to the neutral expression only. Two databases originally captured in the Psychology community are introduced to Computer Vision, to quantitatively demonstrate the influence of expression changes on age estimation, and evaluate the proposed framework and corresponding methods for cross-expression age estimation.

Can We Minimize the Influence Due to Gender and Race in Age Estimation

Automatic human age estimation has attracted a great deal of interest in the past few years. Although many advancements have been made by researchers, there are still many challenges: such as age estimation across different image acquisition methods, different expressions, gender and races. The influence due to race and gender seems to be the most common issue, because collecting a large amount of face images with comprehensive racial diversities seems impractical. The performance will degrade when estimating face images of races that differ from the training set. In this work, we present a new scheme to mitigate the influences of race and gender in the problem of age estimation. Our system will contribute a robust solution to solve the problem of age estimation across races and genders. This study is essential for developing a practical age estimation system (with mixture of races and gender.) To evaluate the performance of the proposed algorithm, we run comprehensive experiments on one widely used big database - MORPH-II, which contains more than 55, 000 images. On an average, an improvement of more than 20% has been achieved using the proposed scheme.

Leveraging appearance and geometry for kinship verification

Kinship verification has become a very active topic recently. Many kinship verification algorithms have been proposed, however, many problems still need to be solved, such as how to locate familial traits of two individuals with kinship and how to make use of facial geometry information to verify kinship relation. To solve these problems, we propose one feature matching scheme to locate familial traits between two facial images. We also advocate a method using geometry information for kinship verification. This approach achieves more than 17% improvement compared with the state of the art. In addition, preliminary results show that the proposed problems can be accurately addressed when fusing appearance and geometry feature.

Age estimation via unsupervised neural networks

In this work, we investigate an unsupervised neural network framework for the problem of facial image age estimation. Unlike previous approaches in age estimation where a predefined feature extraction framework is used, the features used in this work are directly learned from the data. A single-layer convolutional neural network and recursive convolutional neural networks are used to extract features from an image. Manifold learning scheme is incorporated in the framework, which maps the features into the discriminative subspace. Furthermore, several popular regression and classification methods are evaluated using this scheme. As far as we know, this is the first work where an unsupervised neural network has been introduced to the age estimation problem. We evaluate the proposed scheme on two widely used datasets. The experimental results show that there is a significant improvement compared to the state-of-the-art.

Gender classification of depth images based on shape and texture analysis

Gender classification of depth images is a challenging problem, most research work attempted to use shape information to solve this problem in the past literature. In this work, we propose a new fusion scheme for gender classification using both texture and shape features. A new ensemble scheme is advocated to combine texture and shape feature at the feature level. To evaluate the performance of our algorithm, we measure our scheme on two different datasets. The final classification result is up to 93.7% using five-fold cross validation on the whole FRGCv2 dataset, which is comparable to the classification result obtained using visible imagery.

A New Approach for 2D-3D Heterogeneous Face Recognition

This paper proposes a novel scheme for face recognition from visible images to depth images. In our proposed technique, we adopt Partial Least Square (PLS) to handle correlation mapping between 2D to 3D. A considerable performance improvement is observed compared to using Canonical Correlation Analysis (CCA). To further improve the performance, a fusion scheme based on PLS and CCA is advocated. We evaluate the advocated approach on a popular face dataset-FRGCV2.0. Experimental results demonstrate that the proposed scheme is an effective approach to perform 2D-3D face recognition.

Leveraging multiple cues for recognizing family photos

Social relation analysis via images is a new research area that has attracted much interest recently. As social media usage increases, a wide variety of information can be extracted from the growing number of consumer photos shared online, such as the category of events captured or the relationships between individuals in a given picture. Family is one of the most important units in our society, thus categorizing family photos constitutes an essential step toward image-based social analysis and content-based retrieval of consumer photos. We propose an approach that combines multiple unique and complimentary cues for recognizing family photos. The first cue analyzes the geometric arrangement of people in the photograph, which characterizes scene-level information with efficient yet discriminative capability. The second cue models facial appearance similarities to capture and quantify relevant pairwise relations between individuals in a given photo. The last cue investigates the semantics of the context in which the photo was taken. Experiments on a dataset containing thousands of family and non-family pictures collected from social media indicate that each individual model produces good recognition results. Furthermore, a combined approach incorporating appearance, geometric and semantic features significantly outperforms the state of the art in this domain, achieving 96.7% classification accuracy. A new geometry feature is proposed to capture peoples standing pattern at the scene level.Deep convolutional neural network is incorporated into appearance model to capture facial similarities of the group photo.Semantic information is applied and fused with other information to discriminant two different photo categories.

Leveraging geometry and appearance cues for recognizing family photos

Analyzing social relations through image processing is an active and emerging research topic. Family is the basic unit of a society; recognizing and categorizing family photos is an essential step towards image-based social analysis. In this paper, we propose an approach that leverages a geometric model and an appearance model for family photo detection. The geometric feature captures scene-level information with fast computation and discriminative capability, while appearance model captures and quantifies pairwise relations between individuals in a photo. Each individual model performs well for family photo recognition, and a combined approach with both appearance and geometric features can outperform the state-of-the-art method.