Juha Ylioinas

Age Estimation Using Local Binary Pattern Kernel Density Estimate

We propose a novel kernel method for constructing local binary pattern statistics for facial representation in human age estimation. For age estimation, we make use of the de facto support vector regression technique. The main contributions of our work include (i) evaluation of a pose correction method based on simple image flipping and (ii) a comparison of two local binary pattern based facial representations, namely a spatially enhanced histogram and a novel kernel density estimate. Our single- and cross-database experiments indicate that the kernel density estimate based representation yields better estimation accuracy than the corresponding histogram one, which we regard as a very interesting finding. In overall, the constructed age estimation system provides comparable performance against the state-of-the-art methods. We are using a well-defined evaluation protocol allowing a fair comparison of our results.

Combining contrast information and local binary patterns for gender classification

Recent developments in face analysis showed that local binary patterns (LBP) provide excellent results in representing faces. LBP is by definition a purely gray-scale invariant texture operator, codifying only the facial patterns while ignoring the magnitude of gray level differences (i.e. contrast). However, pattern information is independent of the gray scale, whereas contrast is not. On the other hand, contrast is not affected by rotation, but patterns are, by default. So, these two measures can supplement each other. This paper addresses how well facial images can be described by means of both contrast information and local binary patterns. We investigate a new facial representation which combines both measures and extensively evaluate the proposed representation on the gender classification problem, showing interesting results. Furthermore, we compare our results against those of using Haar-like features and AdaBoost learning, demonstrating improvements with a significant margin.

Relative Camera Pose Estimation Using Convolutional Neural Networks

This paper presents a convolutional neural network based approach for estimating the relative pose between two cameras. The proposed network takes RGB images from both cameras as input and directly produces the relative rotation and translation as output. The system is trained in an end-to-end manner utilising transfer learning from a large scale classification dataset. The introduced approach is compared with widely used local feature based methods (SURF, ORB) and the results indicate a clear improvement over the baseline. In addition, a variant of the proposed architecture containing a spatial pyramid pooling (SPP) layer is evaluated and shown to further improve the performance.

Gender and texture classification

The paper fits the topics of the special issue as it deals with gender classification.A review of 13 recent and popular local binary patterns (LBP) variants is presented.A comparative analysis on two problems (gender and texture classification) is given.Extensive experiments showed that basic LBP generalizes well to different problems.The best results are obtained by BSIF but at the cost of higher computational time. Among very popular local image descriptors which has shown interesting results in extracting soft facial biometric traits is the local binary patterns (LBP). LBP is a gray-scale invariant texture operator which labels the pixels of an image by thresholding the neighborhood of each pixel with the value of the center pixel and considers the result as a binary number. LBP labels can be regarded as local primitives such as curved edges, spots, flat areas etc. These labels or their statistics, most commonly the histogram, can then be used for further image analysis. Due to its discriminative power and computational simplicity, the LBP methodology has already attained an established position in computer vision. LBP is also very flexible: it can be easily adapted to different types of problems and used together with other image descriptors. Since its introduction, LBP has inspired plenty of new methods, thus revealing that texture based region descriptors can be very efficient in representing different images. Nowadays, many LBP variants can be found in the literature. This article reviews 13 variants and provides a comparative analysis on two different problems (gender and texture classification) using benchmark databases. The experiments show that basic LBP provides good results and generalizes well to different problems and hence can be a good starting point when trying to find an optimal variant for a given application. The best results are obtained with BSIF (binarized statistical image features) but at the cost of higher computational time compared to basic LBP. Furthermore, experiments on combining three best performing descriptors are conducted, pointing out useful insight into their complementarity.

Face Recognition Using Smoothed High-Dimensional Representation

Recent studies have underlined the significance of high-dimensional features and their compression for face recognition. Partly motivated by these findings, we propose a novel method for building unsupervised face representations based on binarized descriptors and efficient compression by soft assignment and unsupervised dimensionality reduction. For binarized descriptors, we consider Binarized Statistical Image Features (BSIF) which is a learning based descriptor computing a binary code for each pixel by thresholding the outputs of a linear projection between a local image patch and a set of independent basis vectors estimated from a training data set using independent component analysis. In this work, we propose application specific learning to train a separate BSIF descriptor for each of the local face regions. Then, our method constructs a high-dimensional representation from an input face by collecting histograms of BSIF codes in a blockwise manner. Before dropping the dimension to get a more compressed representation, an important step in the pipeline of our method is soft feature assignment where the region histograms of the binarized codes are smoothed using kernel density estimation achieved by a simple and fast matrix-vector product. In detail, we provide a thorough evaluation on FERET and LFW benchmarks comparing our face representation method to the state-of-the-art in face recognition showing enhanced performance on FERET and promising results on LFW.

Constructing Local Binary Pattern Statistics by Soft Voting

In this paper we propose a novel method for constructing Local Binary Pattern (LBP) statistics for image appearance description. The method is inspired by the kernel density estimation designed for estimating the underlying probability function of a random variable. An essential part of the proposed method is the use of Hamming distance. Compared to the standard LBP histogram statistics where one labeled pixel always contributes to one bin of the histogram, the proposed method exploits a kernel-like similarity function to determine weighted votes contributing several possible pattern types in the statistic. As a result, the method yields a more reliable estimate of the underlying LBP distribution of the given image. In overall, the method is easy to implement and outperforms the standard LBP histogram description in texture classification and in biometrics-related face verification. We demonstrate that the method is extremely potential in problems where the number of pixels is limited. This makes the method very promising, for example, in low-resolution image description and the description of interest regions. Another interesting property of the proposed method is that it can be easily integrated with many existing LBP variants that use label statistics as descriptors.

An In-depth Examination of Local Binary Descriptors in Unconstrained Face Recognition

Automatic face recognition in unconstrained conditions is a difficult task which has recently attained increasing attention. In this domain, face verification methods have significantly improved since the release of the Labeled Faces in the Wild database, but the related problem of face identification, is still lacking considerations, which is partly because of the shortage of representative databases. Only recently, two new datasets called Remote Face and Point-and-Shoot Challenge were published providing appropriate benchmarks for the research community to investigate the problem of face recognition in challenging imaging conditions, in both, verification and identification modes. In this paper we provide an in-depth examination of three local binary description methods in unconstrained face recognition evaluating them on these two recently published datasets. In detail, we investigate three well established methods separately and fusing them at rank- and score-levels. We are using a well-defined evaluation protocol allowing a fair comparison of our results for future examinations.

Face and texture analysis using local descriptors: A comparative analysis

In contrast to global image descriptors which compute features directly from the entire image, local descriptors representing the features in small local image patches have proved to be more effective in real world conditions. This paper considers three recent yet popular local descriptors, namely Local Binary Patterns (LBP), Local Phase Quantization (LPQ) and Binarized Statistical Image Features (BSIF), and provides extensive comparative analysis on two different research problems (gender and texture classification) using benchmark datasets. The three descriptors are analyzed in terms of both classification accuracy and computational costs. Furthermore, experiments on combining these descriptors are provided, pointing out useful insight into their complementarity.

Unsupervised learning of overcomplete face descriptors

The current state-of-the-art indicates that a very discriminative unsupervised face representation can be constructed by encoding overlapping multi-scale face image patches at facial landmarks. If fixed as such, there are even suggestions (albeit subtle) that the underlying features may no longer have as much meaning. In spite of the effectiveness of this strategy, we argue that one may still afford to improve especially at the feature level. In this paper, we investigate the role of overcompleteness in features for building unsupervised face representations. In our approach, we first learn an overcomplete basis from a set of sampled face image patches. Then, we use this basis to produce features that are further encoded using the Bag-of-Features (BoF) approach. Using our method, without an extensive use of facial landmarks, one is able to construct a single-scale representation reaching state-of-the-art performance in face recognition and age estimation following the protocols of LFW, FERET, and Adience benchmarks. Furthermore, we make several interesting findings related, for example, to the positive impact of applying soft feature encoding scheme preceding standard dimensionality reduction. To this end, making the encoding faster, we propose a novel method for approximative soft-assignment which we show to perform better than its hard-assigned counterpart.

Data-driven techniques for smoothing histograms of local binary patterns

Local binary pattern histograms have proved very successful texture descriptors. Despite this success, the description procedure bears some drawbacks that are still lacking solutions in the literature. One of the problems arises when the number of extractable local patterns reduces while their dimension increases rendering the output histogram descriptions sparse and unstable, finally showing up as a reduced recognition rate. A smoothing method based on kernel density estimation was recently proposed as a means to tackle the aforementioned problem. A constituent part of the method is to determine how much to smooth a histogram. Previously, this was solved via trial-and-error in a problem-specific manner. In this paper, the goal is to present data-driven methods to determine this smoothing automatically. In the end, we present unsupervised and supervised methods for the given task and validate their performance with a representative set of local binary pattern variants in texture analysis problems covering material categorization and face recognition. HighlightsThis paper proposes data-driven techniques for smoothing LBP histograms.The proposed smoothing techniques cover unsupervised and supervised variants.The techniques are evaluated on material categorization and face recognition.Histogram smoothing is beneficial especially in small-sample-size scenarios.