Object tracking using SIFT features and mean shift

A scale invariant feature transform (SIFT) based mean shift algorithm is presented for object tracking in real scenarios. SIFT features are used to correspond the region of interests across frames. Meanwhile, mean shift is applied to conduct similarity search via color histograms. The probability distributions from these two measurements are evaluated in an expectation–maximization scheme so as to achieve maximum likelihood estimation of similar regions. This mutual support mechanism can lead to consistent tracking performance if one of the two measurements becomes unstable. Experimental work demonstrates that the proposed mean shift/SIFT strategy improves the tracking performance of the classical mean shift and SIFT tracking algorithms in complicated real scenarios.

Where are they now? Tracking the longitudinal evolution of environmental businesses from the 1990s

In 1991 Bennett published one of the first major publications on the concept of ecopreneurship, business opportunities resulting from the emerging environmental agenda of the late 1980s and early 1990s. Since then a body of literature has developed that explores the idea of the intersection of entrepreneurship with environmentally and socially responsible behaviour. Many of the business cases presented by Bennett represent early adopters of green products, services and emerging eco-markets. Given the current emphasis on the transformation of business practices towards a more sustainable paradigm it is timely to review these 94 early ecopreneurial examples and consider their status two decades on from the original publication. This paper explores the definitions of environmental and social enterprise, and considers the longitudinal survival of these companies and the emerging trends in consolidation and failure of the sampled companies.

2D silhouette and 3D skeletal models for human detection and tracking

In this paper we propose a statistical model for detection and tracking of human silhouette and the corresponding 3D skeletal structure in gait sequences. We follow a point distribution model (PDM) approach using a Principal Component Analysis (PCA). The problem of non-lineal PCA is partially resolved by applying a different PDM depending of pose estimation; frontal, lateral and diagonal, estimated by Fisher's linear discriminant. Additionally, the fitting is carried out by selecting the closest allowable shape from the training set by means of a nearest neighbor classifier. To improve the performance of the model we develop a human gait analysis to take into account temporal dynamic to track the human body. The incorporation of temporal constraints on the model increase reliability and robustness.

Mean field approach for tracking similar objects

In this paper, we consider the problem of tracking similar objects. We show how a mean field approach can be used to deal with interacting targets and we compare it with Markov Chain Monte Carlo (MCMC). Two mean field implementations are presented. The first one is more general and uses particle filtering. We discuss some simplifications of the base algorithm that reduce the computation time. The second one is based on suitable Gaussian approximations of probability densities that lead to a set of self-consistent equations for the means and covariances. These equations give the Kalman solution if there is no interaction. Experiments have been performed on two kinds of sequences. The first kind is composed of a single long sequence of twenty roaming ants and was previously analysed using MCMC. In this case, our mean field algorithms obtain substantially better results. The second kind corresponds to selected sequences of a football match in which the interaction avoids tracker coalescence in situations where independent trackers fail.

Gaussian Approximation for Tracking Occluding and Interacting Targets

In this paper, we show how interacting and occluding targets can be tackled successfully within a Gaussian approximation. For that purpose, we develop a general expansion of the mean and covariance of the posterior and we consider a first order approximation of it. The proposed method differs from EKF in that neither a non-linear dynamical model nor a non-linear measurement vector to state relation have to be defined, so it works with any kind of interaction potential and likelihood. The approach has been tested on three sequences (10400, 2500, and 400 frames each one). The results show that our approach helps to reduce the number of failures without increasing too much the computation time with respect to methods that do not take into account target interactions.

Primed, constant infusion with [H-2(3)]serine allows in vivo kinetic measurement of serine turnover, homocysteine remethylation, and transsulfuration processes in human one-carbon metabolism

Background: One-carbon metabolism involves both mitochondrial and cytosolic forms of folate-dependent enzymes in mammalian cells, but few in vivo data exist to characterize the biochemical processes involved.

Objective: We conducted a stable-isotopic investigation to determine the fates of exogenous serine and serine-derived one carbon units in homocysteine remethylation in hepatic and whole-body metabolism.

Design: A healthy man aged 23 y was administered [2,3,3 H-2(3)]serine and [5,5,5-H-2(3)]leucine by intravenous primed, constant infusion. Serial plasma samples were analyzed to determine the isotopic enrichment of free glycine, serine, leucine, methionine, and cystathionine. VLDL apolipoprotein B-100 served as an index of liver free amino acid labeling.

Results: [H-2(1)]Methionine and [H-2(2)]methionine were labeled through homocysteine remethylation. We propose that [H-2(2)]methionine occurs by remethylation with [H-2(2)]methyl groups (as 5-methyltetrahydrofolate) formed only from cytosolic processing of [H-2(3)]serine, whereas [H-2(1)]methionine is formed with labeled one-carbon units from mitochondrial oxidation of C-3 serine to [H-2(1)]formate to yield cytosolic [H-2(1)]methyl groups. The labeling pattern of cystathionine formed from homocysteine and labeled serine suggests that cystathionine is derived mainly from a serine pool different from that used in apolipoprotein B-100 synthesis.

Conclusions: The appearance of both [H-2(1)]- and [H-2(2)]methionine forms indicates that both cytosolic and mitochondrial metabolism of exogenous serine generates carbon units in vivo for methyl group production and homocysteine remethylation. This study also showed the utility of serine infusion and indicated functional roles of cytosolic and mitochondrial compartments in one-carbon metabolism.

Cavitary melanoma of the ciliary body. A study of eight cases

Purpose: The authors present the unique clinical features of cavitary uveal melanoma. Design: Retrospective chart review. Participants: Eight patients with cavitary uveal melanoma. Main Outcome Measures: The clinical, ultrasonographic, and histopathologic features of eight patients with cavitary melanoma of the ciliary body were studied. Results: In all eyes there was a brown ciliary body mass that blocked transmission of light on trans-scleral transillumination. Ocular ultrasonography revealed a large, single hollow cavity (unilocular 'pseudocyst') in five cases and multiple hollow cavities (multilocular 'pseudocyst') in three cases. The cavity occupied a mean of 55% of the entire mass thickness (range, 31%-79%). In five cases, a basal uveal mass was noted on ultrasonography. Four patients underwent tumor resection; one had enucleation, and three had 1251 radioactive plaque treatment. In the five cases confirmed histopathologically, the cavitation was empty, contained erythrocytes, serous fluid, and/or pigment-laden macrophages. In no case was the cavity lined by necrotic tumor, endothelial cells, or epithelial cells. Conclusion: Ciliary body melanoma can develop an intralesional cavity resembling an intraocular cyst. The presence of a solid mass at the base and a thick wall surrounding the cavity can assist in the differentiation of cavitary melanoma from benign cyst.

Energy expenditure during activity in the American lobster Homarus americanus:Correlations with body acceleration

How animals manage time and expend energy has implications for survivorship. Being able to measure key metabolic costs of animals under natural conditions is therefore an important tool in behavioral ecology. One method for estimating activity-specific metabolic rate is via derived measures of acceleration, often 'overall dynamic body acceleration' (ODBA), recorded by an instrumented acceleration logger. ODBA has been shown to correlate well with rate of oxygen consumption (V ?o) in a range of species during activity in the laboratory. This study devised a method for attaching acceleration loggers to decapod crustaceans and then correlated ODBA against concurrent respirometry readings to assess accelerometry as a proxy for activity-specific energy expenditure in a model species, the American lobster Homarus americanus. Where the instrumented animals exhibited a sufficient range of activity levels, positive linear relationships were found between V ?o and ODBA over 20min periods at a range of ambient temperatures (6, 13 and 20°C). Mixed effect linear models based on these data and morphometrics provided reasonably strong predictive power for estimating activity-specific V ?o from ODBA. These V ?o-ODBA calibrations demonstrate the potential of accelerometry as an effective predictor of behavior-specific metabolic rate of crustaceans in the wild during periods of activity. © 2013 Elsevier Inc.