Andrew P. Blake

Multiscale Categorical Object Recognition Using Contour Fragments

Psychophysical studies show that we can recognize objects using fragments of outline contour alone. This paper proposes a new automatic visual recognition system based only on local contour features, capable of localizing objects in space and scale. The system first builds a class-specific codebook of local fragments of contour using a novel formulation of chamfer matching. These local fragments allow recognition that is robust to within-class variation, pose changes, and articulation. Boosting combines these fragments into a cascaded sliding-window classifier, and mean shift is used to select strong responses as a final set of detection. We show how learning can be performed iteratively on both training and test sets to bootstrap an improved classifier. We compare with other methods based on contour and local descriptors in our detailed evaluation over 17 challenging categories and obtain highly competitive results. The results confirm that contour is indeed a powerful cue for multiscale and multiclass visual object recognition.

Efficiently combining contour and texture cues for object recognition

This paper proposes an efficient fusion of contour and texture cues for image categorization and object detection. Our work confirms and strengthens recent results that combining complementary feature types improves performance. We obtain a similar improvement in accuracy and additionally an improvement in efficiency. We use a boosting algorithm to learn models that use contour and texture features. Our main contributions are (i) the use of dense generic texture features to complement contour fragments, and (ii) a simple feature selection mechanism that includes the computational costs of features in order to learn a run-time efficient model. Our evaluation on 17 challenging and varied object classes confirms that the synergy of the two feature types performs significantly better than either alone, and that computational efficiency is substantially improved using our feature selection mechanism. An investigation of the boosted features shows a fascinating emergent property: the absence of certain textures often contributes towards object detection. Comparison with recent work shows that performance is state of the art.

Macroeconomic Policy : Inflation, Wealth and the Exchange Rate

Part 1 Theory: a new Keynesian framework for macroeconomic policy the linkages between financial weapons and financial targets - a comparative static analysis the dynamics of price stabilization. Part 2 Application: a stock-flow model with model-consistent or regressive expectations macroeconomic policy rules for economic stabilization counterfactual simulation with forward-looking expectations a simulation of the cost-push economy the controlled economy with reformed wages. Part 3 Method: the derivation and use of linear model the design of economic policy rules. Part 4 Conclusion: summary of results and conclusions.

A radial basis function artificial neural network test for ARCH

We propose a test for ARCH that uses a radial basis function artificial neural network. It outperforms alternative neural network tests in a variety of Monte Carlo experiments.

Pure Significance Tests of the Unit Root Hypothesis Against Nonlinear Alternatives

This paper describes artificial neural network based pure significance tests for the unit root hypothesis against nonlinear alternatives. The theoretical properties of the tests are discussed and a Monte Carlo investigation of their small sample properties is undertaken. Copyright 2003 Blackwell Publishing Ltd.

Should the Bank of England be Independent

since monetary policy was effectively determined by the actions of the Bundesbank. Now, with the UKs with drawal from the EMS and no immediate prospect of any return, the question has again returned to the policy agenda. In fact, the arguments used to justify such a delegation of responsibility are closely related to those originally associated with the decision to join the EMS itself; that monetary policy should be controlled by an authority with a strong aversion to inflation and a repu tation for sticking to its announced policy. It has been argued that an independent Bank of England would offer precisely these advantages without compromising the domestic (ie UK) objectives which were overridden by the Bundesbank.

Filtered least squares and measurement error

Li, Maddala, and Rush (1995) proposed a low-pass spectral filter method to estimate cointegrating vectors in small samples. This paper tests the effectiveness of the approach in the presence of measurement error. Two other methods, valid under the assumption of stationarity, are also tested.

Exchange-Rate Targets and Wage Formation

In the context of economic management, with final targets for money GDP, wealth and the foreign exchange reserves, the exchange rate can be treated as an intermediate target. Monetary policy is used to keep the exchange rate close to its target and the target itself adjusts in response to information about the final targets. The target exchange rate can be used with main emphasis on either money GDP or on wealth. The latter is not possible if the wage—price loop is powerful. Two reruns of history are presented. The first uses the exchange rate mainly to look after money GDP with wages following their historical behaviour. The second uses fiscal policy to look after money GDP and the exchange rate to look after national wealth. It requires a reform of wage bargaining to be successful.

Optimality and Taylor Rules

This paper discusses the role of forecasts in the control of inflation. Much has been made of variations on the so-called Taylor rule for inflation control. Forward-looking Taylor rules are reconciled with optimal control using a class of rules described as error-correcting Taylor rules.

STERLING, THE EURO AND THE DOLLAR

In the period of floating exchange rates a distinct dollar-D-mark polarisation was noted, whereby peri ods of weakness for some European currencies relative to the dollar coincided with periods of strength relative to the Deutsche mark and vice versa. The relationship continued to be observed for those currencies such as sterling which remained outside the Exchange Rate Mechanism for most of the period. The observation, al though more an empirical regularity in search of a theory than a model (Buiter et al., 1998) seems reason ably robust; recently the dollar has fallen sharply against sterling while the euro has risen. An analysis of this relationship and its stability is important because it provides a guide as to how the euro might be expected to move against sterling as a part of any adjustment against the US dollar. In turn it sheds some light on the degree to which a dollar depreciation might affect Britain inside the Euro Area.