Learning a Family of Detectors

Object detection and recognition are important problems in computer vision. The challenges of these problems come from the presence of noise, background clutter, large within class variations of the object class and limited training data. In addition, the computational complexity in the recognition process is also a concern in practice. In this thesis, we propose one approach to handle the problem of detecting an object class that exhibits large within-class variations, and a second approach to speed up the classification processes. In the first approach, we show that foreground-background classification (detection) and within-class classification of the foreground class (pose estimation) can be jointly solved with using a multiplicative form of two kernel functions. One kernel measures similarity for foreground-background classification. The other kernel accounts for latent factors that control within-class variation and implicitly enables feature sharing among foreground training samples. For applications where explicit parameterization of the within-class states is unavailable, a nonparametric formulation of the kernel can be constructed with a proper foreground distance/similarity measure. Detector training is accomplished via standard Support Vector Machine learning. The resulting detectors are tuned to specific variations in the foreground class. They also serve to evaluate hypotheses of the foreground state. When the image masks for foreground objects are provided in training, the detectors can also produce object segmentation. Methods for generating a representative sample set of detectors are proposed that can enable efficient detection and tracking. In addition, because individual detectors verify hypotheses of foreground state, they can also be incorporated in a tracking-by-detection frame work to recover foreground state in image sequences. To run the detectors efficiently at the online stage, an input-sensitive speedup strategy is proposed to select the most relevant detectors quickly. The proposed approach is tested on data sets of human hands, vehicles and human faces. On all data sets, the proposed approach achieves improved detection accuracy over the best competing approaches. In the second part of the thesis, we formulate a filter-and-refine scheme to speed up recognition processes. The binary outputs of the weak classifiers in a boosted detector are used to identify a small number of candidate foreground state hypotheses quickly via Hamming distance or weighted Hamming distance. The approach is evaluated in three applications: face recognition on the face recognition grand challenge version 2 data set, hand shape detection and parameter estimation on a hand data set, and vehicle detection and estimation of the view angle on a multi-pose vehicle data set. On all data sets, our approach is at least five times faster than simply evaluating all foreground state hypotheses with virtually no loss in classification accuracy.

3D Hand Pose Reconstruction Using Specialized Mappings

A system for recovering 3D hand pose from monocular color sequences is proposed. The system employs a non-linear supervised learning framework, the specialized mappings architecture (SMA), to map image features to likely 3D hand poses. The SMA's fundamental components are a set of specialized forward mapping functions, and a single feedback matching function. The forward functions are estimated directly from training data, which in our case are examples of hand joint configurations and their corresponding visual features. The joint angle data in the training set is obtained via a CyberGlove, a glove with 22 sensors that monitor the angular motions of the palm and fingers. In training, the visual features are generated using a computer graphics module that renders the hand from arbitrary viewpoints given the 22 joint angles. We test our system both on synthetic sequences and on sequences taken with a color camera. The system automatically detects and tracks both hands of the user, calculates the appropriate features, and estimates the 3D hand joint angles from those features. Results are encouraging given the complexity of the task.

Face Identification by a Cascade of Rejection Classifiers

Nearest neighbor search is commonly employed in face recognition but it does not scale well to large dataset sizes. A strategy to combine rejection classifiers into a cascade for face identification is proposed in this paper. A rejection classifier for a pair of classes is defined to reject at least one of the classes with high confidence. These rejection classifiers are able to share discriminants in feature space and at the same time have high confidence in the rejection decision. In the face identification problem, it is possible that a pair of known individual faces are very dissimilar. It is very unlikely that both of them are close to an unknown face in the feature space. Hence, only one of them needs to be considered. Using a cascade structure of rejection classifiers, the scope of nearest neighbor search can be reduced significantly. Experiments on Face Recognition Grand Challenge (FRGC) version 1 data demonstrate that the proposed method achieves significant speed up and an accuracy comparable with the brute force Nearest Neighbor method. In addition, a graph cut based clustering technique is employed to demonstrate that the pairwise separability of these rejection classifiers is capable of semantic grouping.

An Independent-Connection Model for Traffic Matrices

A common assumption made in traffic matrix (TM) modeling and estimation is independence of a packet's network ingress and egress. We argue that in real IP networks, this assumption should not and does not hold. The fact that most traffic consists of two-way exchanges of packets means that traffic streams flowing in opposite directions at any point in the network are not independent. In this paper we propose a model for traffic matrices based on independence of connections rather than packets. We argue that the independent connection (IC) model is more intuitive, and has a more direct connection to underlying network phenomena than the gravity model. To validate the IC model, we show that it fits real data better than the gravity model and that it works well as a prior in the TM estimation problem. We study the model's parameters empirically and identify useful stability properties. This justifies the use of the simpler versions of the model for TM applications. To illustrate the utility of the model we focus on two such applications: synthetic TM generation and TM estimation. To the best of our knowledge this is the first traffic matrix model that incorporates properties of bidirectional traffic.

A Cortical and Cerebellar Parcellation System for Speech Studies

A neuroanatomical parcellation system is described which encompasses the entire cerebral cortex and the cerebellum. The cortical system modified version of the scheme described by Caviness et al. (1996) and is designed particularly for studies of speech processing. The cerebellum is parcellated into 6 cortical regions of interest (ROIs) and an ROI representing the deep cerebellar nuclei in each hemisphere. The boundaries of each ROI are based on individual anatomical markers that are clearly visible from standard structural MRI acquistions. The system permits averaginh of functional imaging data sets from multiple sujects while accounting for individual anatomical variability. Used in conjuction with region-of-interest analysis techniques such as that described by Nieto-Castanon et al. (2003), the parcellation system provides a more powerful means of analyzing functional data.

Speaker Normalization Using Cortical Strip Maps: A Neural Model for Steady State Vowel Identification

Auditory signals of speech are speaker-dependent, but representations of language meaning are speaker-independent. Such a transformation enables speech to be understood from different speakers. A neural model is presented that performs speaker normalization to generate a pitchindependent representation of speech sounds, while also preserving information about speaker identity. This speaker-invariant representation is categorized into unitized speech items, which input to sequential working memories whose distributed patterns can be categorized, or chunked, into syllable and word representations. The proposed model fits into an emerging model of auditory streaming and speech categorization. The auditory streaming and speaker normalization parts of the model both use multiple strip representations and asymmetric competitive circuits, thereby suggesting that these two circuits arose from similar neural designs. The normalized speech items are rapidly categorized and stably remembered by Adaptive Resonance Theory circuits. Simulations use synthesized steady-state vowels from the Peterson and Barney [J. Acoust. Soc. Am. 24, 175-184 (1952)] vowel database and achieve accuracy rates similar to those achieved by human listeners. These results are compared to behavioral data and other speaker normalization models.

Statistical Properties of Single and Competing Non-Linear Fast-Slow Oscillators in Noise

Statistical properties offast-slow Ellias-Grossberg oscillators are studied in response to deterministic and noisy inputs. Oscillatory responses remain stable in noise due to the slow inhibitory variable, which establishes an adaptation level that centers the oscillatory responses of the fast excitatory variable to deterministic and noisy inputs. Competitive interactions between oscillators improve the stability in noise. Although individual oscillation amplitudes decrease with input amplitude, the average to'tal activity increases with input amplitude, thereby suggesting that oscillator output is evaluated by a slow process at downstream network sites.

The Construction of Arbitrary Stable Dynamics in Non-Linear Neural Networks

In this paper, two methods for constructing systems of ordinary differential equations realizing any fixed finite set of equilibria in any fixed finite dimension are introduced; no spurious equilibria are possible for either method. By using the first method, one can construct a system with the fewest number of equilibria, given a fixed set of attractors. Using a strict Lyapunov function for each of these differential equations, a large class of systems with the same set of equilibria is constructed. A method of fitting these nonlinear systems to trajectories is proposed. In addition, a general method which will produce an arbitrary number of periodic orbits of shapes of arbitrary complexity is also discussed. A more general second method is given to construct a differential equation which converges to a fixed given finite set of equilibria. This technique is much more general in that it allows this set of equilibria to have any of a large class of indices which are consistent with the Morse Inequalities. It is clear that this class is not universal, because there is a large class of additional vector fields with convergent dynamics which cannot be constructed by the above method. The easiest way to see this is to enumerate the set of Morse indices which can be obtained by the above method and compare this class with the class of Morse indices of arbitrary differential equations with convergent dynamics. The former set of indices are a proper subclass of the latter, therefore, the above construction cannot be universal. In general, it is a difficult open problem to construct a specific example of a differential equation with a given fixed set of equilibria, permissible Morse indices, and permissible connections between stable and unstable manifolds. A strict Lyapunov function is given for this second case as well. This strict Lyapunov function as above enables construction of a large class of examples consistent with these more complicated dynamics and indices. The determination of all the basins of attraction in the general case for these systems is also difficult and open.

A Unified Model of Spatiotemporal Processing in the Retina

A computational model of visual processing in the vertebrate retina provides a unified explanation of a range of data previously treated by disparate models. Three results are reported here: the model proposes a functional explanation for the primary feed-forward retinal circuit found in vertebrate retinae, it shows how this retinal circuit combines nonlinear adaptation with the desirable properties of linear processing, and it accounts for the origin of parallel transient (nonlinear) and sustained (linear) visual processing streams as simple variants of the same retinal circuit. The retina, owing to its accessibility and to its fundamental role in the initial transduction of light into neural signals, is among the most extensively studied neural structures in the nervous system. Since the pioneering anatomical work by Ramón y Cajal at the turn of the last century[1], technological advances have abetted detailed descriptions of the physiological, pharmacological, and functional properties of many types of retinal cells. However, the relationship between structure and function in the retina is still poorly understood. This article outlines a computational model developed to address fundamental constraints of biological visual systems. Neurons that process nonnegative input signals-such as retinal illuminance-are subject to an inescapable tradeoff between accurate processing in the spatial and temporal domains. Accurate processing in both domains can be achieved with a model that combines nonlinear mechanisms for temporal and spatial adaptation within three layers of feed-forward processing. The resulting architecture is structurally similar to the feed-forward retinal circuit connecting photoreceptors to retinal ganglion cells through bipolar cells. This similarity suggests that the three-layer structure observed in all vertebrate retinae[2] is a required minimal anatomy for accurate spatiotemporal visual processing. This hypothesis is supported through computer simulations showing that the model's output layer accounts for many properties of retinal ganglion cells[3],[4],[5],[6]. Moreover, the model shows how the retina can extend its dynamic range through nonlinear adaptation while exhibiting seemingly linear behavior in response to a variety of spatiotemporal input stimuli. This property is the basis for the prediction that the same retinal circuit can account for both sustained (X) and transient (Y) cat ganglion cells[7] by simple morphological changes. The ability to generate distinct functional behaviors by simple changes in cell morphology suggests that different functional pathways originating in the retina may have evolved from a unified anatomy designed to cope with the constraints of low-level biological vision.

It Takes Two to Tango: Entrepreneurship and Creativity in Troubled Times – Vietnam 2012

Strikingly, most literature suggests that market competition will push firms to take creativity/innovation seriously as matter of death or survival. Using the data, we examined creativity methods (Napier and Nilsson, 2008; Napier, 2010) in conjunction with three influential cultural values – namely risk tolerance, relationship, and dependence on resources – to assess how they influence decisions of entrepreneurs.The primary objective of this study focuses on perceived values of entrepreneurship and creativity in business conducted within a turbulent environment. Our initial hypothesis is that a typical entrepreneurial process carries with it “creativity-enabling elements.” In a normal situation, when businesses focus more on optimizing their resources for commercial gains, perceptions about values of entrepreneurial creativity are usually vague. However, in difficult times and harsh competition, the difference between survival and failure may be creativity. This paper also examines many previous findings on both entrepreneurship and creativity and suggests a highly possible “organic growth” of creativity in an entrepreneurial environment and reinforcing value of entrepreneurship when creativity power is present. In other words, we see each idea reinforcing the other. We use data from a survey of sample Vietnamese firms during the chaotic economic year 2012 to learn about the ‘entrepreneurshipcreativity nexus.’ A data set of 137 responses qualified for a statistical examination was obtained from an online survey, which started on February 16 and ended May 24, 2012, sent to local entrepreneurs and corporate managers using social networks. The authors employed categorical data analysis (Agresti, 2002; Azen & Walker, 2011). Statistical analyses confirm that for business operation, the creativity and entrepreneurial spirit could hardly be separate; and, this is not only correct with entrepreneurial firm, but also well established companies. The single most important factor before business startup and during early implementation in Vietnam is what we call “connection/relationship.” However, businesspeople are increasingly aware of the need of creativity/innovation. In fact, we suggest that creativity and entrepreneurial spirit cannot be separated in entrepreneurial firms as well as established companies.

Valuing ecosystem services from wetlands restoration in the Mississippi Alluvial Valley

This study assesses the value of restoring forested wetlands via the U.S. government's Wetlands Reserve Program (WRP) in the Mississippi Alluvial Valley by quantifying and monetizing ecosystem services. The three focal services are greenhouse gas (GHG) mitigation, nitrogen mitigation, and waterfowl recreation. Site- and region-level measurements of these ecosystem services are combined with process models to quantify their production on agricultural land, which serves as the baseline, and on restored wetlands. We adjust and transform these measures into per-hectare, valuation-ready units and monetize them with prices from emerging ecosystem markets and the environmental economics literature. By valuing three of the many ecosystem services produced, we generate lower bound estimates for the total ecosystem value of the wetlands restoration. Social welfare value is found to be between $1435 and $1486/ha/year, with GHG mitigation valued in the range of $171 to $222, nitrogen mitigation at $1248, and waterfowl recreation at $16. Limited to existing markets, the estimate for annual market value is merely $70/ha, but when fully accounting for potential markets, this estimate rises to $1035/ha. The estimated social value surpasses the public expenditure or social cost of wetlands restoration in only 1 year, indicating that the return on public investment is very attractive for the WRP. Moreover, the potential market value is substantially greater than landowner opportunity costs, showing that payments to private landowners to restore wetlands could also be profitable for individual landowners. © 2009 Elsevier B.V.

The role of B cells in solid organ transplantation.

The role of antibodies in chronic injury to organ transplants has been suggested for many years, but recently emphasized by new data. We have observed that when immunosuppressive potency decreases either by intentional weaning of maintenance agents or due to homeostatic repopulation after immune cell depletion, the threshold of B cell activation may be lowered. In human transplant recipients the result may be donor-specific antibody, C4d+ injury, and chronic rejection. This scenario has precise parallels in a rhesus monkey renal allograft model in which T cells are depleted with CD3 immunotoxin, or in a CD52-T cell transgenic mouse model using alemtuzumab to deplete T cells. Such animal models may be useful for the testing of therapeutic strategies to prevent DSA. We agree with others who suggest that weaning of immunosuppression may place transplant recipients at risk of chronic antibody-mediated rejection, and that strategies to prevent this scenario are needed if we are to improve long-term graft and patient outcomes in transplantation. We believe that animal models will play a crucial role in defining the pathophysiology of antibody-mediated rejection and in developing effective therapies to prevent graft injury. Two such animal models are described herein.

Perceived cannabis use norms and cannabis use among adolescents in the United States.

Due to changes in cannabis policies, concerns about cannabis use (CU) in adolescents have increased. The population of nonwhite groups is growing quickly in the United States. We examined perceived CU norms and their association with CU and CU disorder (CUD) for White, Black, Hispanic, Native-American, Asian-American, Native Hawaiian/Pacific Islander (NH/PI), and mixed-race adolescents. Data were from adolescents (12-17 years) in the 2004-2012 National Surveys on Drug Use and Health (N = 163,837). Substance use and CUD were assessed by computer-assisted, self-interviewing methods. Blacks, Hispanics, Native-Americans, and mixed-race adolescents had greater odds of past-year CU and CUD than Whites. Among past-year cannabis users (CUs), Hispanics and Native-Americans had greater odds of having a CUD than Whites. Asian-Americans had the highest prevalence of perceived parental or close friends' CU disapproval. Native-Americans and mixed-race adolescents had lower odds than Whites of perceiving CU disapproval from parents or close friends. In adjusted analyses, adolescent's disapproval of CU, as well as perceived disapproval by parents or close friends, were associated with a decreased odds of CU in each racial/ethnic group, except for NHs/PIs. Adolescent's disapproval of CU was associated with a decreased odds of CUD among CUs for Whites (personal, parental, and close friends' disapproval), Hispanics (personal, parental, and close friends' disapproval), and mixed-race adolescents (personal, close friends' disapproval). Racial/ethnic differences in adolescent CU prevalence were somewhat consistent with adolescents' reports of CU norm patterns. Longitudinal research on CU health effects should oversample nonwhite adolescents to assure an adequate sample for analysis and reporting.

Conocimientos y criterios para la evaluación de la calidad y concentración energética de los forrajes ensilados

El valor energético de los alimentos se determina mediante experimentos con animales pero que consumen gran cantidad de recursos económicos y humanos. La evaluación energética en la práctica rutinaria de laboratorio depende de la existencia de métodos de estimación basadas sobre parámetros químicos o valores de digestibilidad in vitro capaces de predecir adecuadamente la digestibilidad in vivo. El objetivo general fue generar conocimientos y criterios que mejoren la evaluación de la calidad de los forrajes ensilados y contribuir al desarrollo de ecuaciones de predicción de la digestibilidad de la materia seca (DMS) y de la energía metabolizable (EM) de los forrajes ensilados, utilizando como método de referencia a la digestibilidad obtenida in vivo con ovinos. Las fracciones analíticas como (a) FDN(mo), LIG/FDA y almidón resultaron fundamentales para el desarrollo de las ecuaciones pero otras variables derivadas de la PG fueron necesarias para reducir la magnitud de los errores de la predicción. Las tasas de PG y los tiempos en que éstas ocurrieron fueron de mayor utilidad que las PGA (i.e. Tmax, µ50 y T50). Los parámetros de la cinética de PG de los residuos insolubles en agua y detergente neutro (i.e. RIA y RIDN) permitieron mejorar la precisión de la predicción de DMS y EM reduciendo los errores de los modelos hasta 21.3 g/kg MS y 0.10 Mcal/kg MS, respectivamente (i.e DMS=450 + 10.8*RIA(Tmax) + o.6*(Lig/FDA), R²(aj) =0.94; EM=2.97 - 0.022*RIA(T50 - 0.007*(FDA) + 0.008*(Lig/FDA), R²(aj) =0.89). Las ecuaciones para predecir DFDN presentaron pobres niveles de ajuste (i.e. RDS mayora 40 g/kg FDN). Sería necesario continuar explorando las razones de la variabilidad en los resultados obtenidos. La aplicación de las ecuaciones propuestas requiere la utilización de las fracciones analíticas en base seca corregida por liofilización así como la incubación de los ensilajes en fresco pero molidos con hielo seco.

Rotating-frame nuclear magnetic relaxation of spins diffusing on a disordered lattice: a Monte Carlo model

The rotating-frame nuclear magnetic relaxation rate of spins diffusing on a disordered lattice has been calculated by Monte Carlo methods. The disorder includes not only variation in the distances between neighbouring spin sites but also variation in the hopping rate associated with each site. The presence of the disorder, particularly the hopping rate disorder, causes changes in the time-dependent spin correlation functions which translate into asymmetry in the characteristic peak in the temperature dependence of the dipolar relaxation rate. The results may be used to deduce the average hopping rate from the relaxation but the effect is not sufficiently marked to enable the distribution of the hopping rates to be evaluated. The distribution, which is a measure of the degree of disorder, is the more interesting feature and it has been possible to show from the calculation that measurements of the relaxation rate as a function of the strength of the radiofrequency spin-locking magnetic field can lead to an evaluation of its width. Some experimental data on an amorphous metal - hydrogen alloy are reported which demonstrate the feasibility of this novel approach to rotating-frame relaxation in disordered materials.