Multiple Condensation filters for road detection and tracking

In this paper, we present a novel and robust road tracking system for vision-based personal navigation. Novelty of the work includes the use of multiple Condensation filters to track the road of arbitrary shape and automatic switching between trackers according to road conditions. The approach allows the road to be represented as a simple hyperbola. It also supports the representation of the road as a sequence of connected arcs/segments so that information from a digital map can be integrated into tracking. The parameters of the hyperbola road model are estimated using multiple vanishing points located in image strips. The road tracking method is robust in dealing with complex road shapes, background clutters, shadows, and road markings. Experiments using real videos demonstrate the robustness of our approach.     

An improved approximation to the precision of fixed effects from restricted maximum likelihood

An approximate small sample variance estimator for fixed effects from the multivariate normal linear model, together with appropriate inference tools based on a scaled F pivot, is now well established in practice and there is a growing literature on its properties in a variety of settings. Although effective under linear covariance structures, there are examples of nonlinear structures for which it does not perform as well. The cause of this problem is shown to be a missing term in the underlying Taylor series expansion which accommodates the bias in the estimators of the parameters of the covariance structure. The form of this missing term is derived, and then used to adjust the small sample variance estimator. The behaviour of the resulting estimator is explored in terms of invariance under transformation of the covariance parameters and also using a simulation study. It is seen to perform successfully in the way predicted from its derivation.     

The Meteor metric for automatic evaluation of machine translation

The Meteor Automatic Metric for Machine Translation evaluation, originally developed and released in 2004, was designed with the explicit goal of producing sentence-level scores which correlate well with human judgments of translation quality. Several key design decisions were incorporated into Meteor in support of this goal. In contrast with IBM’s Bleu, which uses only precision-based features, Meteor uses and emphasizes recall in addition to precision, a property that has been confirmed by several metrics as being critical for high correlation with human judgments. Meteor also addresses the problem of reference translation variability by utilizing flexible word matching, allowing for morphological variants and synonyms to be taken into account as legitimate correspondences. Furthermore, the feature ingredients within Meteor are parameterized, allowing for the tuning of the metric’s free parameters in search of values that result in optimal correlation with human judgments. Optimal parameters can be separately tuned for different types of human judgments and for different languages. We discuss the initial design of the Meteor metric, subsequent improvements, and performance in several independent evaluations in recent years.     

On the classification performance of TAN and general Bayesian networks

Over a decade ago, Friedman et al. introduced the Tree Augmented Naïve Bayes (TAN) classifier, with experiments indicating that it significantly outperformed Naïve Bayes (NB) in terms of classification accuracy, whereas general Bayesian network (GBN) classifiers performed no better than NB. This paper challenges those claims, using a careful experimental analysis to show that GBN classifiers significantly outperform NB on datasets analyzed, and are comparable to TAN performance. It is found that the poor performance reported by Friedman et al. are not attributable to the GBN per se, but rather to their use of simple empirical frequencies to estimate GBN parameters, whereas basic parameter smoothing (used in their TAN analyses but not their GBN analyses) improves GBN performance significantly. It is concluded that, while GBN classifiers may have some limitations, they deserve greater attention, particularly in domains where insight into classification decisions, as well as good accuracy, is required.     

On designing safe and flexible embedded DSLs with Java 5

Although embedded DSLs are widely used in both functional and dynamic languages they are yet to be widely accepted in the Java community. The best available examples are projects like jMock and Hibernate Criteria Query. In this paper we introduce the design of three novel practical embedded DSLs that we designed to showcase the power of Java language and specifically the changes introduced in Java 5. One of those DSLs is fully implemented as an open source project and is gathering momentum in the community. We proceed to discuss the approaches that allow designing safe and flexible DSLs by extracting generic patterns where possible and describing more specific idioms where necessary. It is our goal that this paper would serve as a starting point for someone designing an embedded DSL that takes full advantage of the Java 5 features.     

Constructive semantics for instantaneous reactions

This paper presents some results towards a game-theoretic account of the constructive semantics of step responses for synchronous languages, providing a coherent semantic framework encompassing both non-deterministic Statecharts (as per Pnueli & Shalev) and deterministic esterel. In particular, it is shown that esterel arises from a finiteness condition on strategies whereas Statecharts permits infinite games. Beyond giving a novel and unifying account of these concrete languages the paper sketches a general theory for obtaining different notions of constructive responses in terms of winning conditions for finite and infinite games and their characterisation as maximal post-fixed points of functions in directed complete lattices of intensional truth-values.     

Discrete retardance second harmonic generation ellipsometry

A new instrument was constructed to perform discrete retardance nonlinear optical ellipsometry (DR-NOE). The focus of the design was to perform second harmonic generation NOE while maximizing sample and application flexibility and minimizing data acquisition time. The discrete retardance configuration results in relatively simple computational algorithms for performing nonlinear optical ellipsometric analysis. NOE analysis of a disperse red 19 monolayer yielded results that were consistent with previously reported values for the same surface system, but with significantly reduced acquisition times.     

A simple Hebbian/anti-Hebbian network learns the sparse, independent components of natural images

Slightly modified versions of an early Hebbian/anti-Hebbian neural network are shown to be capable of extracting the sparse, independent linear components of a prefiltered natural image set. An explanation for this capability in terms of a coupling between two hypothetical networks is presented. The simple networks presented here provide alternative, biologically plausible mechanisms for sparse, factorial coding in early primate vision.