We introduce a paradigm for understanding physical scenes without human annotations. At the core of our system is a physical world representation that is first recovered by a perception module and then utilized by physics and graphics engines. During training, the perception module and the generative models learn by visual de-animation — interpreting and reconstructing the visual information st...

In this paper, we describe and situate the TUPELO system for data mapping in relational databases. Automating the discovery of mappings between structured data sources is a long standing and important problem in data management. Starting from user provided example instances of the source and target schemas, TUPELO approaches mapping discovery as search within the transformation space of these i...

This paper presents an improved method of training for the unvoiced filter that comprises an excitation model, within the framework of parametric speech synthesis based on hidden Markov models. The conventional approach calculates the unvoiced filter response from the differential signal of the residual and voiced excitation estimate. The differential signal, however, includes the error generat...

At least 90% of trials are extended by at least 6 weeks because investigators fail to enroll patients on schedule (Ledford, 2011). It is therefore important at trial design-time to have good insight in how the choice of the eligibility criteria affects the recruitment rate. Based on that insight, trial designers can then adjust the eligibility criteria in order to ensure realistic recruiting ra...

A previous publication has described a method of pairwise 3D surface correspondence for the automated generation of landmarks on a set of examples from a class of shape [3]. In this paper we describe a set of improved algorithms which give more accurate and more robust results. We show how the pairwise corresponder can be used in a extension of an existing framework for establishing dense corre...

We propose a method for learning causal relations within high-dimensional tensor data as they are typically recorded in non-experimental databases. The method allows the simultaneous inclusion of numerous dimensions within the data analysis such as samples, time and domain variables construed as tensors. In such tensor data we exploit and integrate non-Gaussian models and tensor analytic algori...

Agent based modelling has been widely accepted as a promising tool for urban planning purposes thanks to its capability to provide sophisticated insights into the social behaviours and the interdependencies that characterise urban systems. In this paper, we report on an agent based model, called TransMob, which explicitly simulates the mutual dynamics between demographic evolution, transport de...

The Fukushima Daiichi disaster 11 March, 2011 is considered the largest nuclear accident since the 1986 Chernobyl disaster and has been rated at level 7 on the International Nuclear Event Scale. As different radioactive materials have different effects to human body, it is important to know the types of nuclides and their levels of concentration from the recorded mixture of radiations to take n...

Time series techniques are applied to Ghara-Aghaj flow records, in order to generate forecast values of the mean monthly river flows. The study of data and its correlogram shows the effect of seasonality and provide no evidence of trend. The autoregressive models of order one and two (AR1, AR2), moving average model of order one and ARMA (1,1) model are fitted to the stationary series, where th...

The application of porous bio-nanocomposites polymer has greatly increased in the treatment of boneabnormalities and bone fracture. Therefore, predicting the mechanical properties of these bio-nanocompositesare very important prior to their fabrication. Investigation of mechanical properties like (elasticmodulus and hardness) is very costly and time-consuming in experimental t...