Recovering facial pose with the EM algorithm

This paper describes how 3D facial pose may be estimated by fitting a template to 2D feature locations. The fitting process is realised as projecting the control points of a 3D template onto the 2D feature locations under orthographic projection. The parameters of the orthographic projection are iteratively estimated using the EM algorithm. The method is evaluated on both contrived data with known ground-truth together with some more naturalistic imagery. These experiments reveal that under favourable conditions the algorithm can estimate facial pitch to within 3°.     

Toward a taxonomy of human performance.

Discusses general problems in the development of taxonomic systems for describing human tasks and performance. Alternative approaches and provisional classification schemes are presented. Specific techniques of measurement and scaling, applicable to certain task classification systems, are described and their reliability evaluated. Attempts to evaluate these systems are summarized, and attempts to apply them to several areas of human performance research (e.g., studies of drug effects, learning procedures, alcohol, and vigilance) are examined. A series of studies linking task characteristics with ability requirements is described. Some of this research is considered encouraging, in that the generalizability of data on performance increases when certain classification systems are used to describe the tasks utilized in such research. (44 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Some perspectives for industrial psychology.

Discusses several matters deemed important to industrial psychologists. It is suggested that, given the intangible character of psychological variables, it would be fruitful to obtain the ideas of ordinary people about the variables that are significant in occupational behavior. Industrial psychologists ought to study organizations as "individuals" rather than just regarding them as social environments. The use of simulated organizations (e.g., mathematical models) would facilitate such investigations. Industrial psychologists should consider the differences among people to be quantitative rather than qualitative. Consequently, they should not devote their time to investigating differences among arbitrary types of people, but rather should direct their attention to the quantitative variables (e.g., social factors) which underlie those qualitatively different categories. The role and nature of theory and the impermanence of facts which emerge from empirical studies are also discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reactivity of heat-treated coals in carbon dioxide at 900 °C

Reactivities of sixteen 40 × 100 (U.S.) mesh U.S. coals charred to 1000 °C were measured in carbon dioxide at 900 °C. Chars derived from coals with less than 80% carbon, on a dry-ash-free basis, were the most reactive. These chars also gave the widest spread in reactivity. Plots of inorganic element content in the chars versus reactivity showed that magnesium and calcium are important to char reactivity. Six coals were acid-washed with hydrochloric acid and four coals were further demineralized with hydrofluoric acid. Most acid-treated coals showed a decrease in reactivity; but two coals of high rank increased in reactivity. This increase in reactivity is attributed to the creation of additional porosity as a result of mineral matter removal and thus a reduction in resistance to carbon dioxide diffusion to reactive sites. Two demineralized and two original coals were divided into four size ranges and chars were produced from each size of each coal. Gasification rates increased monotonically with decreasing particle size reacted.     

Operational Amplifier Circuit for Linearizing Temperature Readings from Thermistors

An operational amplifier circuit is presented that will provide a linear relation between output voltage and temperature from 10°C to 50°C using a nonlinear thermistor as the temperature transducer. The source of the linearity is discussed and a general method for determining circuit values is described.     

Fugacity-based indoor residential pesticide fate model

Dermal and nondietary pathways are possibly important for exposure to pesticides used in residences. Limited data have been collected on pesticide concentrations in residential air and surfaces following application. Models may be useful for interpreting these data and to make predictions about concentrations in the home for other pesticides based on chemical properties. We present a dynamic mass-balance compartment model based on fugacity principles. The model includes air (both gas phase and aerosols), carpet, smooth flooring, and walls as model compartments. Six size fractions of particulate matter with different fate and transport properties are included. We determine the compartmental fugacity capacity and mass-transfer rate coefficients between compartments. We compare model results to chlorpyrifos air and carpet measurements from an independent study. For a comparison, we run the same simulation for diazinon and permethrin. We quantify the effect of parameter uncertainty and model uncertainties related to the source release rate and conduct a sensitivity analysis to determine which parameters contribute most to output uncertainty. In the model comparison to chlorpyrifos measurements, the model results are of the same order of magnitude as measured values but tend to overpredict the measured data, thus indicating the need for a better understanding of emissions from treated surfaces.