Face recognition from video

In spite of over two decades of intense research, illumination and pose invariance remain prohibitively challenging aspects of face recognition for most practical applications. The objective of this work is to recognize faces using video sequences both for training and recognition input, in a realistic, unconstrained setup in which lighting, pose and user motion pattern have a wide variability and face images are of low resolution. In particular there are three areas of novelty: (i) we show how a photometric model of image formation can be combined with a statistical model of generic face appearance variation, learnt offline, to generalize in the presence of extreme illumination changes; (ii) we use the smoothness of geodesically local appearance manifold structure and a robust same-identity likelihood to achieve invariance to unseen head poses; and (iii) we introduce an accurate video sequence "reillumination" algorithm to achieve robustness to face motion patterns in video. We describe a fully automatic recognition system based on the proposed method and an extensive evaluation on 171 individuals and over 1300 video sequences with extreme illumination, pose and head motion variation. On this challenging data set our system consistently demonstrated a nearly perfect recognition rate (over 99.7%), significantly outperforming state-of-the-art commercial software and methods from the literature. © Springer-Verlag Berlin Heidelberg 2006.

Natural versus artificial scene classification by ordering discrete fourier power spectra

Holistic representations of natural scenes is an effective and powerful source of information for semantic classification and analysis of arbitrary images. Recently, the frequency domain has been successfully exploited to holistically encode the content of natural scenes in order to obtain a robust representation for scene classification. In this paper, we present a new approach to naturalness classification of scenes using frequency domain. The proposed method is based on the ordering of the Discrete Fourier Power Spectra. Features extracted from this ordering are shown sufficient to build a robust holistic representation for Natural vs. Artificial scene classification. Experiments show that the proposed frequency domain method matches the accuracy of other state-of-the-art solutions. © 2008 Springer Berlin Heidelberg.

Gaussian mixture modeling with Gaussian process latent variable models

Density modeling is notoriously difficult for high dimensional data. One approach to the problem is to search for a lower dimensional manifold which captures the main characteristics of the data. Recently, the Gaussian Process Latent Variable Model (GPLVM) has successfully been used to find low dimensional manifolds in a variety of complex data. The GPLVM consists of a set of points in a low dimensional latent space, and a stochastic map to the observed space. We show how it can be interpreted as a density model in the observed space. However, the GPLVM is not trained as a density model and therefore yields bad density estimates. We propose a new training strategy and obtain improved generalisation performance and better density estimates in comparative evaluations on several benchmark data sets. © 2010 Springer-Verlag.

Beyond dataset bias: Multi-task unaligned shared knowledge transfer

Many visual datasets are traditionally used to analyze the performance of different learning techniques. The evaluation is usually done within each dataset, therefore it is questionable if such results are a reliable indicator of true generalization ability. We propose here an algorithm to exploit the existing data resources when learning on a new multiclass problem. Our main idea is to identify an image representation that decomposes orthogonally into two subspaces: a part specific to each dataset, and a part generic to, and therefore shared between, all the considered source sets. This allows us to use the generic representation as un-biased reference knowledge for a novel classification task. By casting the method in the multi-view setting, we also make it possible to use different features for different databases. We call the algorithm MUST, Multitask Unaligned Shared knowledge Transfer. Through extensive experiments on five public datasets, we show that MUST consistently improves the cross-datasets generalization performance. © 2013 Springer-Verlag.

Rationales of holonic manufacturing systems in leather industry

This paper presents an insight into leather manufacturing processes, depicting peculiarities and challenges faced by leather industry. An analysis of this industry reveals the need for a new approach to optimize the productivity of leather processing operations, ensure consistent quality of leather, mitigate the adverse health effects in tannery workers exposed to chemicals and comply with environmental regulation. Holonic manufacturing systems (HMS) paradigm represent a bottom-up distributed approach that provides stability, adaptability, efficient use of resources and a plug and operate functionality to the manufacturing system. A vision of how HMS might operate in a tannery is illustrated presenting the rationales behind its application in this industry. © 2013 Springer-Verlag.

Product intelligence in warehouse management: A case study

The need for more flexible, adaptable and customer-oriented warehouse operations has been increasingly identified as an important issue by today's warehouse companies due to the rapidly changing preferences of the customers that use their services. Motivated by manufacturing and other logistics operations, in this paper we argue on the potential application of product intelligence in warehouse operations as an approach that can help warehouse companies address these issues. We discuss the opportunities of such an approach using a real example of a third-party-logistics warehouse company and we present the benefits it can bring in their warehouse management systems. © 2013 Springer-Verlag.

Automatic construction and natural-language description of nonparametric regression models

Copyright © 2014, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. This paper presents the beginnings of an automatic statistician, focusing on regression problems. Our system explores an open-ended space of statistical models to discover a good explanation of a data set, and then produces a detailed report with figures and natural- language text. Our approach treats unknown regression functions non- parametrically using Gaussian processes, which has two important consequences. First, Gaussian processes can model functions in terms of high-level properties (e.g. smoothness, trends, periodicity, changepoints). Taken together with the compositional structure of our language of models this allows us to automatically describe functions in simple terms. Second, the use of flexible nonparametric models and a rich language for composing them in an open-ended manner also results in state- of-the-art extrapolation performance evaluated over 13 real time series data sets from various domains.

Anisotropy preserving DTI processing

Statistical analysis of diffusion tensor imaging (DTI) data requires a computational framework that is both numerically tractable (to account for the high dimensional nature of the data) and geometric (to account for the nonlinear nature of diffusion tensors). Building upon earlier studies exploiting a Riemannian framework to address these challenges, the present paper proposes a novel metric and an accompanying computational framework for DTI data processing. The proposed approach grounds the signal processing operations in interpolating curves. Well-chosen interpolating curves are shown to provide a computational framework that is at the same time tractable and information relevant for DTI processing. In addition, and in contrast to earlier methods, it provides an interpolation method which preserves anisotropy, a central information carried by diffusion tensor data. © 2013 Springer Science+Business Media New York.

An overview of the ILSP unit selection text-to-speech synthesis system

This paper presents an overview of the Text-to-Speech synthesis system developed at the Institute for Language and Speech Processing (ILSP). It focuses on the key issues regarding the design of the system components. The system currently fully supports three languages (Greek, English, Bulgarian) and is designed in such a way to be as language and speaker independent as possible. Also, experimental results are presented which show that the system produces high quality synthetic speech in terms of naturalness and intelligibility. The system was recently ranked among the first three systems worldwide in terms of achieved quality for the English language, at the international Blizzard Challenge 2013 workshop. © 2014 Springer International Publishing.

Computational information geometry and its applications

Digitization is the main feature of modern Information Science. Conjoining the digits and the coordinates, the relation between Information Science and high-dimensional space is consanguineous, and the information issues are transformed to the geometry problems in some high-dimensional spaces. From this basic idea, we propose Computational Information Geometry (CIG) to make information analysis and processing. Two kinds of applications of CIG are given, which are blurred image restoration and pattern recognition. Experimental results are satisfying. And in this paper, how to combine with groups of simple operators in some 2D planes to implement the geometrical computations in high-dimensional space is also introduced. Lots of the algorithms have been realized using software.

The structure and function of neurons with variable nonlinear transfer function

One novel neuron with variable nonlinear transfer function is firstly proposed, It could also be called as subsection transfer function neuron. With different transfer function components, by virtue of multi-thresholded, the variable transfer function neuron switch on among different nonlinear excitated state. And the comparison of output's transfer characteristics between it and single-thresholded neuron will be illustrated, with some practical application experiments on Bi-level logic operation, at last the simple comparison with conventional BP, RBF, and even DBF NN is taken to expect the development foreground on the variable neuron.. The novel nonlinear transfer function neuron could implement the random nonlinear mapping relationship between input layer and output layer, which could make variable transfer function neuron have one much wider applications on lots of reseach realm such as function approximation pattern recognition data compress and so on.

incremental learning of triadic plsa for collaborative filtering

Beijing University of Technology (BJUT); Beijing Municipal Lab of Brain Informatics; Chinese Society of Radiology; National Natural Science Foundation of China (NSFC); State Administration of Foreign Experts Affairs

Evaluation of a combined wavelet and a combined principal component analysis classification system for BCG diagnostic problem

Heart disease is one of the main factor causing death in the developed countries. Over several decades, variety of electronic and computer technology have been developed to assist clinical practices for cardiac performance monitoring and heart disease diagnosis. Among these methods, Ballistocardiography (BCG) has an interesting feature that no electrodes are needed to be attached to the body during the measurement. Thus, it is provides a potential application to asses the patients heart condition in the home. In this paper, a comparison is made for two neural networks based BCG signal classification models. One system uses a principal component analysis (PCA) method, and the other a discrete wavelet transform, to reduce the input dimensionality. It is indicated that the combined wavelet transform and neural network has a more reliable performance than the combined PCA and neural network system. Moreover, the wavelet transform requires no prior knowledge of the statistical distribution of data samples and the computation complexity and training time are reduced.

三维物体识别及姿态测定的推理系统

本文介绍了三维物体识别及姿态测定的一种新技术，从物体空间域模型出发，通过约束推理及几何推理，在物体三维信息部分给定的条件下，推断预测图象模型，并通过实测的图象数据反馈，推断出隐含在图象中未给定的三维信息，最终实现三维物体识别及姿态测定。整个系统在ＶＩＣＯＭ机上用Ｃ语言完成。

地震数据波动方程反演方法研究

Geophysical inversion is a theory that transforms the observation data into corresponding geophysical models. The goal of seismic inversion is not only wave velocity models, but also the fine structures and dynamic process of interior of the earth, expanding to more parameters such as density, aeolotropism, viscosity and so on. As is known to all, Inversion theory is divided to linear and non-linear inversion theories. In rencent 40 years linear inversion theory has formed into a complete and systematic theory and found extensive applications in practice. While there are still many urgent problems to be solved in non-linear inversion theory and practice. Based on wave equation, this dissertation has been mainly involved in the theoretical research of several non-linear inversion methods: waveform inversion, traveltime inversion and the joint inversion about two methods. The objective of gradient waveform inversion is to find a geologic model, thus synthetic seismograms generated by this geologic model are best fitted to observed seismograms. Contrasting with other inverse methods, waveform inversion uses all characteristics of waveform and has high resolution capacity. But waveform inversion is an interface by interface method. An artificial parameter limit should be provided in each inversion iteration. In addition, waveform information will tend to get stuck in local minima if the starting model is too far from the actual model. Based on velocity scanning in traditional seismic data processing, a layer-by-layer waveform inversion method is developed in this dissertation to deal with weaknesses of waveform inversion. Wave equation is used to calculate the traveltime and derivative (perturbation of traveltime with respect to velocity) in wave-equation traveltime inversion (WT). Unlike traditional ray-based travetime inversion, WT has many advantages. No ray tracing or traveltime picking and no high frequency assumption is necessary and good result can be got while starting model is far from real model. But, comparing with waveform inversion, WT has low resolution. Waveform inversion and WT have complementary advantages and similar algorithm, which proves that the joint inversion is a better inversion method. And another key point which this dissertation emphasizes is how to give fullest play to their complementary advantages on the premise of no increase of storage spaces and amount of calculation. Numerical tests are implemented to prove the feasibility of inversion methods mentioned above in this dissertation. Especially for gradient waveform inversion, field data are inversed. This field data are acquired by our group in Wali park and Shunyi district. Real data processing shows there are many problems for waveform inversion to deal with real data. The matching of synthetic seismograms with observed seismograms and noise cancellation are two primary problems. In conclusion, on the foundation of the former experiences, this dissertation has implemented waveform inversions on the basis of acoustic wave equation and elastic wave equation, traveltime inversion on the basis of acoustic wave equation and traditional combined waveform traveltime inversion. Besides the traditional analysis of inversion theory, there are two innovations: layer by layer inversion of seimic reflection data inversion and rapid method for acoustic wave-equation joint inversion.