Local inter-session variability modelling for object classification

Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset.

Within-session variability modelling for factor analysis speaker verification

This work presents an extended Joint Factor Analysis model including explicit modelling of unwanted within-session variability. The goals of the proposed extended JFA model are to improve verification performance with short utterances by compensating for the effects of limited or imbalanced phonetic coverage, and to produce a flexible JFA model that is effective over a wide range of utterance lengths without adjusting model parameters such as retraining session subspaces. Experimental results on the 2006 NIST SRE corpus demonstrate the flexibility of the proposed model by providing competitive results over a wide range of utterance lengths without retraining and also yielding modest improvements in a number of conditions over current state-of-the-art.

Applying SVMs and weight-based factor analysis to unsupervised adaptation for speaker verification

This paper presents an extended study on the implementation of support vector machine(SVM) based speaker verification in systems that employ continuous progressive model adaptation using the weight-based factor analysis model. The weight-based factor analysis model compensates for session variations in unsupervised scenarios by incorporating trial confidence measures in the general statistics used in the inter-session variability modelling process. Employing weight-based factor analysis in Gaussian mixture models (GMM) was recently found to provide significant performance gains to unsupervised classification. Further improvements in performance were found through the integration of SVM-based classification in the system by means of GMM supervectors. This study focuses particularly on the way in which a client is represented in the SVM kernel space using single and multiple target supervectors. Experimental results indicate that training client SVMs using a single target supervector maximises performance while exhibiting a certain robustness to the inclusion of impostor training data in the model. Furthermore, the inclusion of low-scoring target trials in the adaptation process is investigated where they were found to significantly aid performance.

Bi-modal biometric authentication on mobile phones in challenging conditions

This paper examines the issue of face, speaker and bi-modal authentication in mobile environments when there is significant condition mismatch. We introduce this mismatch by enrolling client models on high quality biometric samples obtained on a laptop computer and authenticating them on lower quality biometric samples acquired with a mobile phone. To perform these experiments we develop three novel authentication protocols for the large publicly available MOBIO database. We evaluate state-of-the-art face, speaker and bi-modal authentication techniques and show that inter-session variability modelling using Gaussian mixture models provides a consistently robust system for face, speaker and bi-modal authentication. It is also shown that multi-algorithm fusion provides a consistent performance improvement for face, speaker and bi-modal authentication. Using this bi-modal multi-algorithm system we derive a state-of-the-art authentication system that obtains a half total error rate of 6.3% and 1.9% for Female and Male trials, respectively.

Robust face clustering for real-world data

This thesis has investigated how to cluster a large number of faces within a multi-media corpus in the presence of large session variation. Quality metrics are used to select the best faces to represent a sequence of faces; and session variation modelling improves clustering performance in the presence of wide variations across videos. Findings from this thesis contribute to improving the performance of both face verification systems and the fully automated clustering of faces from a large video corpus.

Inter-cycle variability of ignition delay in an ethanol fumigated common rail diesel engine

An experimental study has been performed to investigate the ignition delay of a modern heavy-duty common-rail diesel engine run with fumigated ethanol substitutions up to 40% on an energy basis. The ignition delay was determined through the use of statistical modelling in a Bayesian framework this framework allows for the accurate determination of the start of combustion from single consecutive cycles and does not require any differentiation of the in-cylinder pressure signal. At full load the ignition delay has been shown to decrease with increasing ethanol substitutions and evidence of combustion with high ethanol substitutions prior to diesel injection have also been shown experimentally and by modelling. Whereas, at half load increasing ethanol substitutions have increased the ignition delay. A threshold absolute air to fuel ratio (mole basis) of above ~110 for consistent operation has been determined from the inter-cycle variability of the ignition delay, a result that agrees well with previous research of other in-cylinder parameters and further highlights the correlation between the air to fuel ratio and inter-cycle variability. Numerical modelling to investigate the sensitivity of ethanol combustion has also been performed. It has been shown that ethanol combustion is sensitive to the initial air temperature around the feasible operating conditions of the engine. Moreover, a negative temperature coefficient region of approximately 900{1050 K (the approximate temperature at fuel injection) has been shown with for n-heptane and n-heptane/ethanol blends in the numerical modelling. A consequence of this is that the dominate effect influencing the ignition delay under increasing ethanol substitutions may rather be from an increase in chemical reactions and not from in-cylinder temperature. Further investigation revealed that the chemical reactions at low ethanol substitutions are different compared to the high (> 20%) ethanol substitutions.

Inter-cycle variability of in-cylinder pressure parameters in an ethanol fumigated common rail diesel engine

The effects of ethanol fumigation on the inter-cycle variability of key in-cylinder pressure parameters in a modern common rail diesel engine have been investigated. Specifically, maximum rate of pressure rise, peak pressure, peak pressure timing and ignition delay were investigated. A new methodology for investigating the start of combustion was also proposed and demonstrated—which is particularly useful with noisy in-cylinder pressure data as it can have a significant effect on the calculation of an accurate net rate of heat release indicator diagram. Inter-cycle variability has been traditionally investigated using the coefficient of variation. However, deeper insight into engine operation is given by presenting the results as kernel density estimates; hence, allowing investigation of otherwise unnoticed phenomena, including: multi-modal and skewed behaviour. This study has found that operation of a common rail diesel engine with high ethanol substitutions (>20% at full load, >30% at three quarter load) results in a significant reduction in ignition delay. Further, this study also concluded that if the engine is operated with absolute air to fuel ratios (mole basis) less than 80, the inter-cycle variability is substantially increased compared to normal operation.

In-cylinder pressure and inter-cycle variability analysis for a compression ignition engine : Bayesian approaches

This thesis introduced Bayesian statistics as an analysis technique to isolate resonant frequency information in in-cylinder pressure signals taken from internal combustion engines. Applications of these techniques are relevant to engine design (performance and noise), energy conservation (fuel consumption) and alternative fuel evaluation. The use of Bayesian statistics, over traditional techniques, allowed for a more in-depth investigation into previously difficult to isolate engine parameters on a cycle-by-cycle basis. Specifically, these techniques facilitated the determination of the start of pre-mixed and diffusion combustion and for the in-cylinder temperature profile to be resolved on individual consecutive engine cycles. Dr Bodisco further showed the utility of the Bayesian analysis techniques by applying them to in-cylinder pressure signals taken from a compression ignition engine run with fumigated ethanol.

The influence of fatty acid methyl ester profiles on inter-cycle variability in a heavy duty compression ignition engine

With the advent of alternative fuels, such as biodiesels and related blends, it is important to develop an understanding of their effects on inter-cycle variability which, in turn, influences engine performance as well as its emission. Using four methanol trans-esterified biomass fuels of differing carbon chain length and degree of unsaturation, this paper provides insight into the effect that alternative fuels have on inter-cycle variability. The experiments were conducted with a heavy-duty Cummins, turbo-charged, common-rail compression ignition engine. Combustion performance is reported in terms of the following key in-cylinder parameters: indicated mean effective pressure (IMEP), net heat release rate (NHRR), standard deviation of variability (StDev), coefficient of variation (CoV), peak pressure, peak pressure timing and maximum rate of pressure rise. A link is also established between the cyclic variability and oxygen ratio, which is a good indicator of stoichiometry. The results show that the fatty acid structures did not have a significant effect on injection timing, injection duration, injection pressure, StDev of IMEP, or the timing of peak motoring and combustion pressures. However, a significant effect was noted on the premixed and diffusion combustion proportions, combustion peak pressure and maximum rate of pressure rise. Additionally, the boost pressure, IMEP and combustion peak pressure were found to be directly correlated to the oxygen ratio. The emission of particles positively correlates with oxygen content in the fuel as well as in the air-fuel mixture resulting in a higher total number of particles per unit of mass.

Improving out-domain PLDA speaker verification using unsupervised inter-dataset variability compensation approach

Experimental studies have found that when the state-of-the-art probabilistic linear discriminant analysis (PLDA) speaker verification systems are trained using out-domain data, it significantly affects speaker verification performance due to the mismatch between development data and evaluation data. To overcome this problem we propose a novel unsupervised inter dataset variability (IDV) compensation approach to compensate the dataset mismatch. IDV-compensated PLDA system achieves over 10% relative improvement in EER values over out-domain PLDA system by effectively compensating the mismatch between in-domain and out-domain data.

Automatic speaker recognition under adverse conditions

Speaker verification is the process of verifying the identity of a person by analysing their speech. There are several important applications for automatic speaker verification (ASV) technology including suspect identification, tracking terrorists and detecting a person’s presence at a remote location in the surveillance domain, as well as person authentication for phone banking and credit card transactions in the private sector. Telephones and telephony networks provide a natural medium for these applications. The aim of this work is to improve the usefulness of ASV technology for practical applications in the presence of adverse conditions. In a telephony environment, background noise, handset mismatch, channel distortions, room acoustics and restrictions on the available testing and training data are common sources of errors for ASV systems. Two research themes were pursued to overcome these adverse conditions: Modelling mismatch and modelling uncertainty. To directly address the performance degradation incurred through mismatched conditions it was proposed to directly model this mismatch. Feature mapping was evaluated for combating handset mismatch and was extended through the use of a blind clustering algorithm to remove the need for accurate handset labels for the training data. Mismatch modelling was then generalised by explicitly modelling the session conditions as a constrained offset of the speaker model means. This session variability modelling approach enabled the modelling of arbitrary sources of mismatch, including handset type, and halved the error rates in many cases. Methods to model the uncertainty in speaker model estimates and verification scores were developed to address the difficulties of limited training and testing data. The Bayes factor was introduced to account for the uncertainty of the speaker model estimates in testing by applying Bayesian theory to the verification criterion, with improved performance in matched conditions. Modelling the uncertainty in the verification score itself met with significant success. Estimating a confidence interval for the "true" verification score enabled an order of magnitude reduction in the average quantity of speech required to make a confident verification decision based on a threshold. The confidence measures developed in this work may also have significant applications for forensic speaker verification tasks.

PLDA based speaker verification with weighted LDA techniques

This paper investigates the use of the dimensionality-reduction techniques weighted linear discriminant analysis (WLDA), and weighted median fisher discriminant analysis (WMFD), before probabilistic linear discriminant analysis (PLDA) modeling for the purpose of improving speaker verification performance in the presence of high inter-session variability. Recently it was shown that WLDA techniques can provide improvement over traditional linear discriminant analysis (LDA) for channel compensation in i-vector based speaker verification systems. We show in this paper that the speaker discriminative information that is available in the distance between pair of speakers clustered in the development i-vector space can also be exploited in heavy-tailed PLDA modeling by using the weighted discriminant approaches prior to PLDA modeling. Based upon the results presented within this paper using the NIST 2008 Speaker Recognition Evaluation dataset, we believe that WLDA and WMFD projections before PLDA modeling can provide an improved approach when compared to uncompensated PLDA modeling for i-vector based speaker verification systems.

Improving automatic speaker verification using SVM techniques

Automatic recognition of people is an active field of research with important forensic and security applications. In these applications, it is not always possible for the subject to be in close proximity to the system. Voice represents a human behavioural trait which can be used to recognise people in such situations. Automatic Speaker Verification (ASV) is the process of verifying a persons identity through the analysis of their speech and enables recognition of a subject at a distance over a telephone channel { wired or wireless. A significant amount of research has focussed on the application of Gaussian mixture model (GMM) techniques to speaker verification systems providing state-of-the-art performance. GMM's are a type of generative classifier trained to model the probability distribution of the features used to represent a speaker. Recently introduced to the field of ASV research is the support vector machine (SVM). An SVM is a discriminative classifier requiring examples from both positive and negative classes to train a speaker model. The SVM is based on margin maximisation whereby a hyperplane attempts to separate classes in a high dimensional space. SVMs applied to the task of speaker verification have shown high potential, particularly when used to complement current GMM-based techniques in hybrid systems. This work aims to improve the performance of ASV systems using novel and innovative SVM-based techniques. Research was divided into three main themes: session variability compensation for SVMs; unsupervised model adaptation; and impostor dataset selection. The first theme investigated the differences between the GMM and SVM domains for the modelling of session variability | an aspect crucial for robust speaker verification. Techniques developed to improve the robustness of GMMbased classification were shown to bring about similar benefits to discriminative SVM classification through their integration in the hybrid GMM mean supervector SVM classifier. Further, the domains for the modelling of session variation were contrasted to find a number of common factors, however, the SVM-domain consistently provided marginally better session variation compensation. Minimal complementary information was found between the techniques due to the similarities in how they achieved their objectives. The second theme saw the proposal of a novel model for the purpose of session variation compensation in ASV systems. Continuous progressive model adaptation attempts to improve speaker models by retraining them after exploiting all encountered test utterances during normal use of the system. The introduction of the weight-based factor analysis model provided significant performance improvements of over 60% in an unsupervised scenario. SVM-based classification was then integrated into the progressive system providing further benefits in performance over the GMM counterpart. Analysis demonstrated that SVMs also hold several beneficial characteristics to the task of unsupervised model adaptation prompting further research in the area. In pursuing the final theme, an innovative background dataset selection technique was developed. This technique selects the most appropriate subset of examples from a large and diverse set of candidate impostor observations for use as the SVM background by exploiting the SVM training process. This selection was performed on a per-observation basis so as to overcome the shortcoming of the traditional heuristic-based approach to dataset selection. Results demonstrate the approach to provide performance improvements over both the use of the complete candidate dataset and the best heuristically-selected dataset whilst being only a fraction of the size. The refined dataset was also shown to generalise well to unseen corpora and be highly applicable to the selection of impostor cohorts required in alternate techniques for speaker verification.

Extending the task of diarization to speaker attribution

In this paper we extend the concept of speaker annotation within a single-recording, or speaker diarization, to a collection wide approach we call speaker attribution. Accordingly, speaker attribution is the task of clustering expectantly homogenous intersession clusters obtained using diarization according to common cross-recording identities. The result of attribution is a collection of spoken audio across multiple recordings attributed to speaker identities. In this paper, an attribution system is proposed using mean-only MAP adaptation of a combined-gender UBM to model clusters from a perfect diarization system, as well as a JFA-based system with session variability compensation. The normalized cross-likelihood ratio is calculated for each pair of clusters to construct an attribution matrix and the complete linkage algorithm is employed to conduct clustering of the inter-session clusters. A matched cluster purity and coverage of 87.1% was obtained on the NIST 2008 SRE corpus.