Showing papers by "San Jose State University published in 2019"

Seven HCI Grand Challenges

Abstract: This article aims to investigate the Grand Challenges which arise in the current and emerging landscape of rapid technological evolution towards more intelligent interactive technologies, coupled w...

CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification

Abstract: Urban traffic optimization using traffic cameras as sensors is driving the need to advance state-of-the-art multi-target multi-camera (MTMC) tracking. This work introduces CityFlow, a city-scale traffic camera dataset consisting of more than 3 hours of synchronized HD videos from 40 cameras across 10 intersections, with the longest distance between two simultaneous cameras being 2.5 km. To the best of our knowledge, CityFlow is the largest-scale dataset in terms of spatial coverage and the number of cameras/videos in an urban environment. The dataset contains more than 200K annotated bounding boxes covering a wide range of scenes, viewing angles, vehicle models, and urban traffic flow conditions. Camera geometry and calibration information are provided to aid spatio-temporal analysis. In addition, a subset of the benchmark is made available for the task of image-based vehicle re-identification (ReID). We conducted an extensive experimental evaluation of baselines/state-of-the-art approaches in MTMC tracking, multi-target single-camera (MTSC) tracking, object detection, and image-based ReID on this dataset, analyzing the impact of different network architectures, loss functions, spatio-temporal models and their combinations on task effectiveness. An evaluation server is launched with the release of our benchmark at the 2019 AI City Challenge (this https URL) that allows researchers to compare the performance of their newest techniques. We expect this dataset to catalyze research in this field, propel the state-of-the-art forward, and lead to deployed traffic optimization(s) in the real world.

CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification

Abstract: Urban traffic optimization using traffic cameras as sensors is driving the need to advance state-of-the-art multi-target multi-camera (MTMC) tracking. This work introduces CityFlow, a city-scale traffic camera dataset consisting of more than 3 hours of synchronized HD videos from 40 cameras across 10 intersections, with the longest distance between two simultaneous cameras being 2.5 km. To the best of our knowledge, CityFlow is the largest-scale dataset in terms of spatial coverage and the number of cameras/videos in an urban environment. The dataset contains more than 200K annotated bounding boxes covering a wide range of scenes, viewing angles, vehicle models, and urban traffic flow conditions. Camera geometry and calibration information are provided to aid spatio-temporal analysis. In addition, a subset of the benchmark is made available for the task of image-based vehicle re-identification (ReID). We conducted an extensive experimental evaluation of baselines/state-of-the-art approaches in MTMC tracking, multi-target single-camera (MTSC) tracking, object detection, and image-based ReID on this dataset, analyzing the impact of different network architectures, loss functions, spatio-temporal models and their combinations on task effectiveness. An evaluation server is launched with the release of our benchmark at the 2019 AI City Challenge (https://www.aicitychallenge.org/) that allows researchers to compare the performance of their newest techniques. We expect this dataset to catalyze research in this field, propel the state-of-the-art forward, and lead to deployed traffic optimization(s) in the real world.

Techno–ecological synergies of solar energy for global sustainability

Abstract: The strategic engineering of solar energy technologies—from individual rooftop modules to large solar energy power plants—can confer significant synergistic outcomes across industrial and ecological boundaries. Here, we propose techno–ecological synergy (TES), a framework for engineering mutually beneficial relationships between technological and ecological systems, as an approach to augment the sustainability of solar energy across a diverse suite of recipient environments, including land, food, water, and built-up systems. We provide a conceptual model and framework to describe 16 TESs of solar energy and characterize 20 potential techno–ecological synergistic outcomes of their use. For each solar energy TES, we also introduce metrics and illustrative assessments to demonstrate techno–ecological potential across multiple dimensions. The numerous applications of TES to solar energy technologies are unique among energy systems and represent a powerful frontier in sustainable engineering to minimize unintended consequences on nature associated with a rapid energy transition. Managing the interactions and impacts of scaled-up solar energy production will require understanding of the relationships between technological and ecological systems. This Perspective proposes a framework that could help engineer beneficial outcomes from an energy transition.

Machine learning in electronic-quantum-matter imaging experiments.

Abstract: For centuries, the scientific discovery process has been based on systematic human observation and analysis of natural phenomena1. Today, however, automated instrumentation and large-scale data acquisition are generating datasets of such large volume and complexity as to defy conventional scientific methodology. Radically different scientific approaches are needed, and machine learning (ML) shows great promise for research fields such as materials science2–5. Given the success of ML in the analysis of synthetic data representing electronic quantum matter (EQM)6–16, the next challenge is to apply this approach to experimental data—for example, to the arrays of complex electronic-structure images17 obtained from atomic-scale visualization of EQM. Here we report the development and training of a suite of artificial neural networks (ANNs) designed to recognize different types of order hidden in such EQM image arrays. These ANNs are used to analyse an archive of experimentally derived EQM image arrays from carrier-doped copper oxide Mott insulators. In these noisy and complex data, the ANNs discover the existence of a lattice-commensurate, four-unit-cell periodic, translational-symmetry-breaking EQM state. Further, the ANNs determine that this state is unidirectional, revealing a coincident nematic EQM state. Strong-coupling theories of electronic liquid crystals18,19 are consistent with these observations. A machine-learning approach is used to train artificial neural networks to analyse experimental scanning tunnelling microscopy image arrays of quantum materials.

Stock Price Prediction Using News Sentiment Analysis

Abstract: Predicting stock market prices has been a topic of interest among both analysts and researchers for a long time. Stock prices are hard to predict because of their high volatile nature which depends on diverse political and economic factors, change of leadership, investor sentiment, and many other factors. Predicting stock prices based on either historical data or textual information alone has proven to be insufficient. Existing studies in sentiment analysis have found that there is a strong correlation between the movement of stock prices and the publication of news articles. Several sentiment analysis studies have been attempted at various levels using algorithms such as support vector machines, naive Bayes regression, and deep learning. The accuracy of deep learning algorithms depends upon the amount of training data provided. However, the amount of textual data collected and analyzed during the past studies has been insufficient and thus has resulted in predictions with low accuracy. In our paper, we improve the accuracy of stock price predictions by gathering a large amount of time series data and analyzing it in relation to related news articles, using deep learning models. The dataset we have gathered includes daily stock prices for S&P500 companies for five years, along with more than 265,000 financial news articles related to these companies. Given the large size of the dataset, we use cloud computing as an invaluable resource for training prediction models and performing inference for a given stock in real time.

Formation of ultra-diffuse galaxies in the field and in galaxy groups

Abstract: We study ultra-diffuse galaxies (UDGs) in zoom in cosmological simulations, seeking the origin of UDGs in the field versus galaxy groups. We find that while field UDGs arise from dwarfs in a characteristic mass range by multiple episodes of supernova feedback (Di Cintio et al.), group UDGs may also form by tidal puffing up and they become quiescent by ram-pressure stripping. The field and group UDGs share similar properties, independent of distance from the group centre. Their dark-matter haloes have ordinary spin parameters and centrally dominant dark-matter cores. Their stellar components tend to have a prolate shape with a Sersic index n -1 but no significant rotation. Ram pressure removes the gas from the group UDGs when they are at pericentre, quenching star formation in them and making them redder. This generates a colour/star-formation-rate gradient with distance from the centre of the dense environment, as observed in clusters. We find that -20 per cent of the field UDGs that fall into a massive halo survive as satellite UDGs. In addition, normal field dwarfs on highly eccentric orbits can become UDGs near pericentre due to tidal puffing up, contributing about half of the group-UDG population. We interpret our findings using simple toy models, showing that gas stripping is mostly due to ram pressure rather than tides. We estimate that the energy deposited by tides in the bound component of a satellite over one orbit can cause significant puffing up provided that the orbit is sufficiently eccentric. We caution that while the simulations produce UDGs that match the observations, they under-produce the more compact dwarfs in the same mass range, possibly because of the high threshold for star formation or the strong feedback.

Spin transport in a Mott insulator of ultracold fermions

Abstract: Strongly correlated materials are expected to feature unconventional transport properties, such that charge, spin, and heat conduction are potentially independent probes of the dynamics. In contrast to charge transport, the measurement of spin transport in such materials is highly challenging. We observed spin conduction and diffusion in a system of ultracold fermionic atoms that realizes the half-filled Fermi-Hubbard model. For strong interactions, spin diffusion is driven by super-exchange and doublon-hole–assisted tunneling, and strongly violates the quantum limit of charge diffusion. The technique developed in this work can be extended to finite doping, which can shed light on the complex interplay between spin and charge in the Hubbard model.

Spatially Resolved Stellar Kinematics of the Ultra-diffuse Galaxy Dragonfly 44. I. Observations, Kinematics, and Cold Dark Matter Halo Fits

Abstract: We present spatially resolved stellar kinematics of the well-studied ultra-diffuse galaxy (UDG) Dragonfly 44, as determined from 25.3 hr of observations with the Keck Cosmic Web Imager. The luminosity-weighted dispersion within the half-light radius is σ_(1/2) = 33^(+3)_(−3) km s^(−1), lower than what we had inferred before from a DEIMOS spectrum in the Hα region. There is no evidence for rotation, with V_(max)/ σ < 0.12 (90% confidence) along the major axis, in possible conflict with models where UDGs are the high-spin tail of the normal dwarf galaxy distribution. The spatially averaged line profile is more peaked than a Gaussian, with Gauss–Hermite coefficient h_4 = 0.13 ± 0.05. The mass-to-light ratio (M/L) within the effective radius is (M_(dyn)/L_I)(

Mycoprotein: The Future of Nutritious Nonmeat Protein, a Symposium Review.

Abstract: Mycoprotein is an alternative, nutritious protein source with a meat-like texture made from Fusarium venenatum, a naturally occurring fungus Its unique method of production yields a significantly reduced carbon and water footprint relative to beef and chicken Mycoprotein, sold as Quorn, is consumed in 17 countries, including the United States In line with current dietary guidelines, mycoprotein is high in protein and fiber, and low in fat, cholesterol, sodium, and sugar Mycoprotein may help maintain healthy blood cholesterol levels, promote muscle synthesis, control glucose and insulin levels, and increase satiety It is possible that some susceptible consumers will become sensitized, and subsequently develop a specific allergy However, a systematic evidence review indicates that incidence of allergic reactions remains exceptionally low Mycoprotein's nutritional, health, and environmental benefits affirms its role in a healthful diet Future research that focuses on the long-term clinical benefits of consuming a diet containing mycoprotein is warranted

Still Missing Dark Matter: KCWI High-Resolution Stellar Kinematics of NGC1052-DF2

Abstract: The velocity dispersion of the ultra diffuse galaxy NGC1052-DF2 was found to be km s−1, much lower than expected from the stellar mass–halo mass relation and nearly identical to the expected value from the stellar mass alone. This result was based on the radial velocities of 10 luminous globular clusters that were assumed to be associated with the galaxy. A more precise measurement is possible from high-resolution spectroscopy of the diffuse stellar light. Here we present an integrated spectrum of the diffuse light of NGC1052-DF2 obtained with the Keck Cosmic Web Imager (KCWI), with an instrumental resolution of σ instr ≈ 12 km s−1. The systemic velocity of the galaxy is v sys = 1805 ± 1.1 km s−1, in very good agreement with the average velocity of the globular clusters ( km s−1). There is no evidence for rotation within the KCWI field of view. We find a stellar velocity dispersion of km s−1, consistent with the dispersion that was derived from the globular clusters. The implied dynamical mass within the half-light radius r 1/2 = 2.7 kpc is M dyn = (1.3 ± 0.8) × 108 M ⊙, similar to the stellar mass within that radius (M stars = (1.0 ± 0.2) × 108 M ⊙). With this confirmation of the low velocity dispersion of NGC1052-DF2, the most urgent question is whether this "missing dark matter problem" is unique to this galaxy or applies more widely.

Methods, availability, and applications of PM2.5 exposure estimates derived from ground measurements, satellite, and atmospheric models.

Abstract: Fine particulate matter (PM2.5) is a well-established risk factor for public health. To support both health risk assessment and epidemiological studies, data are needed on spatial and temporal patt...

The 2019 AI City Challenge.

Abstract: The AI City Challenge has been created to accelerate intelligent video analysis that helps make cities smarter and safer. With millions of traffic video cameras acting as sensors around the world, there is a significant opportunity for real-time and batch analysis of these videos to provide actionable insights. These insights will benefit a wide variety of agencies, from traffic control to public safety. The 2019 AI City Challenge is the third annual edition in the AI City Challenge series with significant growing attention and participation. AI City Challenge 2019 enabled 334 academic and industrial research teams from 44 countries to solve real-world problems using real city-scale traffic camera video data. The Challenge was launched with three tracks. Track 1 addressed city-scale multi-camera vehicle tracking, Track 2 addressed city-scale vehicle re-identification, and Track 3 addressed traffic anomaly detection. Each track was chosen in consultation with departments of transportation focusing on problems of greatest public value. With the largest available dataset for such tasks, and ground truth for each track, the 2019 AI City Challenge received 129 submissions from 96 individuals teams (there were 22, 84, 23 team submissions from Tracks 1, 2, and 3 respectively). Participation in this challenge has grown five-fold this year as tasks have become more relevant to traffic optimization and challenging to the computer vision community. Results observed strongly underline the value AI brings to city-scale video analysis for traffic optimization.

Maximizing the intersection of human health and the health of the environment with regard to the amount and type of protein produced and consumed in the United States.

Abstract: This review utilizes current national dietary guidelines and published databases to evaluate the impacts of reasonable shifts in the amount and type of protein intake in the United States on the intersection of human and environmental health. The established scientific basis and recommendations for protein intake as described in the US Dietary Reference Intakes are reviewed. Data on food availability from both the US Department of Agriculture and the Food and Agriculture Organization of the United Nations and data on consumption from the National Health and Nutrition Examination Survey are used to examine estimates of current US protein consumption. Greenhouse gas (carbon dioxide equivalents, CO2eq) and blue and green water impacts of US protein consumption resulting from US agricultural practices were obtained from previously published meta-analyses. A 25% decrease in protein intake paired with a 25% shift from animal food to plant food protein intake-from an 85:15 ratio to a 60:40 ratio-would best align protein intake with national dietary recommendations while simultaneously resulting in 40% fewer CO2eq emissions and 10% less consumptive water use. The modeling of this strategy suggests a savings of 129 billion kilograms of CO2eq and 3.1 trillion gallons of water relative to current consumption.

The Fire and Smoke Model Evaluation Experiment-A Plan for Integrated, Large Fire-Atmosphere Field Campaigns.

Abstract: The Fire and Smoke Model Evaluation Experiment (FASMEE) is designed to collect integrated observations from large wildland fires and provide evaluation datasets for new models and operational systems. Wildland fire, smoke dispersion, and atmospheric chemistry models have become more sophisticated, and next-generation operational models will require evaluation datasets that are coordinated and comprehensive for their evaluation and advancement. Integrated measurements are required, including ground-based observations of fuels and fire behavior, estimates of fire-emitted heat and emissions fluxes, and observations of near-source micrometeorology, plume properties, smoke dispersion, and atmospheric chemistry. To address these requirements the FASMEE campaign design includes a study plan to guide the suite of required measurements in forested sites representative of many prescribed burning programs in the southeastern United States and increasingly common high-intensity fires in the western United States. Here we provide an overview of the proposed experiment and recommendations for key measurements. The FASMEE study provides a template for additional large-scale experimental campaigns to advance fire science and operational fire and smoke models.

Overhauling Ocean Spatial Planning to Improve Marine Megafauna Conservation

Abstract: Tracking data have led to evidence-based conservation of marine megafauna, but a disconnect remains between the many 1000s of individual animals that have been tracked and the use of these data in conservation and management actions. Furthermore, the focus of most conservation efforts is within Exclusive Economic Zones despite the ability of these species to move 1000s of kilometers across multiple national jurisdictions. To assist the goal of the United Nations General Assembly’s recent effort to negotiate a global treaty to conserve biodiversity on the high seas, we propose the development of a new frontier in dynamic marine spatial management. We argue that a global approach combining tracked movements of marine megafauna and human activities at-sea, and using existing and emerging technologies (e.g., through new tracking devices and big data approaches) can be applied to deliver near real-time diagnostics on existing risks and threats to mitigate global risks for marine megafauna. With technology developments over the next decade expected to catalyze the potential to survey marine animals and human activities in ever more detail and at global scales, the development of dynamic predictive tools based on near real-time tracking and environmental data will become crucial to address increasing risks. Such global tools for dynamic spatial and temporal management will, however, require extensive synoptic data updates and will be dependent on a shift to a culture of data sharing and open access. We propose a global mechanism to store and make such data available in near real-time, enabling a holistic view of space use by marine megafauna and humans that would significantly accelerate efforts to mitigate impacts and improve conservation and management of marine megafauna.

A hierarchical approach for creating electrically conductive network structure in polyurethane nanocomposites using a hybrid of graphene nanoplatelets, carbon black and multi-walled carbon nanotubes

Abstract: Hierarchical organization of carbon nanomaterials is the best strategy to combine desirable factors and synergistically impart mechanical and electrical properties to polymers. Here, we investigate the relaxation behavior of carbon nanofillers filled polyurethane (PU) with special reference to particle size and aspect ratio, filler morphology, filler loading to understand the conductive network formation of fillers in the PU matrix. Typically, an addition of 2 wt% hybrid fillers of graphene nanoplatelets (GNPs), conductive carbon black (CB) and multi-walled carbon nanotubes (MWCNTs) in PU at 1:1:2 mass ratio (GCM112-PU2) showed lowest surface resistivity ∼106.8 Ω/sq along with highest improved mechanical properties. Our results demonstrate how hierarchical compositions may function in polymer configurations that are useful for thermal and electrical systems.

Behavior of mice aboard the International Space Station.

Abstract: Interest in space habitation has grown dramatically with planning underway for the first human transit to Mars. Despite a robust history of domestic and international spaceflight research, understanding behavioral adaptation to the space environment for extended durations is scant. Here we report the first detailed behavioral analysis of mice flown in the NASA Rodent Habitat on the International Space Station (ISS). Following 4-day transit from Earth to ISS, video images were acquired on orbit from 16- and 32-week-old female mice. Spaceflown mice engaged in a full range of species-typical behaviors. Physical activity was greater in younger flight mice as compared to identically-housed ground controls, and followed the circadian cycle. Within 9–11 days after launch, younger (but not older), mice began to exhibit distinctive circling or ‘race-tracking’ behavior that evolved into a coordinated group activity. Organized group circling behavior unique to spaceflight may represent stereotyped motor behavior, rewarding effects of physical exercise, or vestibular sensation produced via self-motion. Affording mice the opportunity to grab and run in the RH resembles physical activities that the crew participate in routinely. Our approach yields a useful analog for better understanding human responses to spaceflight, providing the opportunity to assess how physical movement influences responses to microgravity.

Sleep inertia: current insights

Abstract: Sleep inertia, or the grogginess felt upon awakening, is associated with significant cognitive performance decrements that dissipate as time awake increases. This impairment in cognitive performance has been observed in both tightly controlled in-laboratory studies and in real-world scenarios. Further, these decrements in performance are exaggerated by prior sleep loss and the time of day in which a person awakens. This review will examine current insights into the causes of sleep inertia, factors that may positively or negatively influence the degree of sleep inertia, the consequences of sleep inertia both in the laboratory and in real-world settings, and lastly discuss potential countermeasures to lessen the impact of sleep inertia.

Depression Among Older Adults in the United States by Disaggregated Race and Ethnicity.

Abstract: Background and objectives As the population becomes increasingly diverse, it is important to understand the prevalence of depression across a racially and ethnically diverse older population. The purpose of this study was to compare rates of depression by age and disaggregated racial and ethnic groups to inform practitioners and target resource allocation to high risk groups. Research design and methods Data were from the Centers for Medicare and Medicaid Services Health Outcomes Survey, Cohorts 15 and 16, a national and annual survey of a racially diverse group of adults aged 65 and older who participate in Medicare Advantage plans (N = 175,956). Depression was operationalized by the Patient Health Questionnaire-2 (PHQ-2); we estimated a logistic regression model and adjusted standard errors to account for 403 Medicare Advantage Organizations. Results Overall, 10.2% of the sample (n = 17,957) reported a PHQ-2 score of 3 or higher, indicative of a positive screen for depression. After adjusting for covariates, odds of screening positively for depression were higher among participants self-reporting as Mexican (odds ratio [OR] = 1.19), Puerto Rican (OR = 1.46), Cuban (OR = 1.57), another Hispanic/Latino (OR = 1.29), and multiple Hispanic/Latino (OR = 1.84) ethnicities, compared with non-Hispanic whites. Odds were also higher among participants reporting that their race was black/African American (OR = 1.20), Asian Indian (OR = 1.67), Filipino (OR = 1.30), Native Hawaiian/Pacific Islander (OR = 1.82), or two or more races (OR = 1.50), compared with non-Hispanic whites. Discussion and implications Prevalence varied greatly across segments of the population, suggesting that certain racial/ethnic groups are at higher risk than others. These disparities should inform distribution of health care resources; efforts to educate and ameliorate depression should be culturally targeted.