A5002755684- Video Surveillance and Tracking Methods
- Anomaly Detection Techniques and Applications
- Advanced Neural Network Applications
- Traffic Prediction and Management Techniques
- Data Management and Algorithms
- Web Data Mining and Analysis
- Autonomous Vehicle Technology and Safety
- Data Quality and Management
- Time Series Analysis and Forecasting
- Semantic Web and Ontologies
- Energy Load and Power Forecasting
- Human Mobility and Location-Based Analysis
- Text and Document Classification Technologies
- Stock Market Forecasting Methods
- Advanced Text Analysis Techniques
- Advanced Database Systems and Queries
- Hydrological Forecasting Using AI
- Advanced Graph Neural Networks
- Complex Network Analysis Techniques
- Recommender Systems and Techniques
- Smart Cities and Technologies
- Data Mining Algorithms and Applications
- Spam and Phishing Detection
- Advanced Image and Video Retrieval Techniques
- Data-Driven Disease Surveillance
Santa Clara University
2020-2025
San Jose State University
2016-2019
Davidson College
2018
Albany State University
2018
University at Albany, State University of New York
2018
University of Minnesota
2014-2015
Twin Cities Orthopedics
2015
Texas State University
2009-2013
Urban traffic optimization using 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 camera dataset consisting of more than 3 hours synchronized HD videos from 40 across 10 intersections, with longest distance between two simultaneous being 2.5 km. To best our knowledge, CityFlow largest-scale in terms spatial coverage and number cameras/videos an urban environment. The contains 200K...
Predicting stock market prices has been a topic of interest among both analysts and researchers for long time. Stock are hard to predict because their high volatile nature which depends on diverse political economic factors, change leadership, investor sentiment, many other factors. based either historical data or textual information alone proven be insufficient. Existing studies in sentiment analysis have found that there is strong correlation between the movement publication news articles....
The AI City Challenge was created to accelerate intelligent video analysis that helps make cities smarter and safer. Transportation is one of the largest segments can benefit from actionable insights derived data captured by sensors, where computer vision deep learning have shown promise in achieving large-scale practical deployment. 4th annual edition has attracted 315 participating teams across 37 countries, who leverage city-scale real traffic high-quality synthetic compete four challenge...
The NVIDIA AI City Challenge has been created to accelerate intelligent video analysis that helps make cities smarter and safer. With millions of traffic cameras acting as sensors around the world, there is a significant opportunity for real-time batch these videos provide actionable insights. These insights will benefit wide variety agencies, from control public safety. second edition Challenge, being organized CVPR workshop, provided forum more than 70 academic industrial research teams...
The AI City Challenge was created with two goals in mind: (1) pushing the boundaries of research and development intelligent video analysis for smarter cities use cases, (2) assessing tasks where level performance is enough to cause real-world adoption. Transportation a segment ripe such fifth attracted 305 participating teams across 38 countries, who leveraged city-scale real traffic data high-quality synthetic compete five challenge tracks. Track 1 addressed video-based automatic vehicle...
The 6th edition of the AI City Challenge specifically focuses on problems in two domains where there is tremendous unlocked potential at intersection computer vision and artificial intelligence: Intelligent Traffic Systems (ITS), brick mortar retail businesses. four challenge tracks 2022 received participation requests from 254 teams across 27 countries. Track 1 addressed city-scale multi-target multi-camera (MTMC) vehicle tracking. 2 natural-language-based track retrieval. 3 was a brand new...
The AI City Challenge's seventh edition emphasizes two domains at the intersection of computer vision and artificial intelligence - retail business Intelligent Traffic Systems (ITS) that have considerable untapped potential. 2023 challenge had five tracks, which drew a record-breaking number participation requests from 508 teams across 46 countries. Track 1 was brand new track focused on multi-target multi-camera (MTMC) people tracking, where trained evaluated using both real highly...
The retail industry has witnessed a remarkable upswing in the utilization of cutting-edge artificial intelligence and computer vision techniques. Among prominent challenges this domain is development an automated checkout system that can address multifaceted issues arise real-world scenarios, including object occlusion, motion blur, similarity scanned items. In paper, we propose sophisticated deep learning-based framework effectively recognize, localize, track, count products as they...
The rapid recent advancements in the computation ability of everyday computers have made it possible to widely apply deep learning methods analysis traffic surveillance videos. Traffic flow prediction, anomaly detection, vehicle re-identification, and tracking are basic components analysis. Among these applications, or speed estimation, is one most important research topics years. Good solutions this problem could prevent collisions help improve road planning by better estimating transit...
Web image analysis has witnessed an AI renaissance. The ILSVRC benchmark been instrumental in providing a corpus and standardized evaluation. NVIDIA City Challenge is envisioned to provide similar impetus the of video data that helps make cities smarter safer. In its first year, this focused on traffic data. While millions cameras around world capture data, albeit low-quality, very little automated value creation results. Lack labeled trained models can be deployed at edge city fabric,...
This article presents a synthetic distracted driving (SynDD1) dataset for machine learning models to detect and analyze drivers' various behavior different gaze zones. We collected the data in stationary vehicle using three in-vehicle cameras positioned at locations: on dashboard, near rearview mirror, top right-side window corner. The contains two activity types: activities [1], [2], [3], zones [4], [5], [6] each participant type has sets: without appearance blocks with blocks, such as...
Low-latency inference for machine learning models is increasingly becoming a necessary requirement, as these are used in mission-critical applications such autonomous driving, military defense (e.g., target recognition), and network traffic analysis. A widely studied technique to overcome this challenge offload some or all parts of the tasks onto specialized hardware graphic processing units. More recently, offloading programmable devices, interface cards switch, gaining interest from both...
A majority of microbial infections are associated with biofilms. Targeting biofilms is considered an effective strategy to limit virulence while minimizing the development antibiotic resistance. Toward this need, antibiofilm peptides attractive arsenal since they bestowed properties orthogonal small molecule drugs. In work, we developed machine learning models identify distinguishing characteristics known peptides, and mine peptide databases from diverse habitats classify new potential...
The All-Pairs similarity search, or self-similarity join problem, finds all pairs of vectors in a high dimensional sparse dataset with value higher than given threshold. problem has been classically solved using dynamically built inverted index. search time is reduced by early pruning candidates size and value-based bounds on the similarity. In context cosine weighted vectors, leveraging Cauchy-Schwarz inequality, we propose new ℓ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Forecasting time series with extreme events has been a challenging and prevalent research topic, especially when the data are affected by complicated uncertain factors, such as is case in hydrologic prediction. Diverse traditional deep learning models have applied to discover nonlinear relationships recognize complex patterns these types of data. However, existing methods usually ignore negative influence imbalanced data, or severe events, on model training. Moreover, evaluated small number...
In the hydrology field, time series forecasting is crucial for efficient water resource management, improving flood and drought control increasing safety quality of life general population. However, predicting long-term streamflow a complex task due to presence extreme events. It requires capture long-range dependencies modeling rare but important values. Existing approaches often struggle tackle these dual challenges simultaneously. this paper, we specifically delve into issues propose...
The k-nearest neighbor graph is often used as a building block in information retrieval, clustering, online advertising, and recommender systems algorithms. complexity of constructing the exact quadratic on number objects that are compared, most existing methods solve problem approximately. We present L2Knng, an efficient algorithm finds cosine similarity for set sparse high-dimensional objects. Our quickly builds approximate solution to problem, identifying many similar neighbors, then uses...
With the aim of analyzing large-sized multidimensional single-cell datasets, we are describing a method for Cosine-based Tanimoto similarity-refined graph community detection using Leiden's algorithm (CosTaL). As graph-based clustering method, CosTaL transforms cells with high-dimensional features into weighted k-nearest-neighbor (kNN) graph. The represented by vertices graph, while an edge between two in represents close relatedness cells. Specifically, builds exact kNN cosine similarity...
Abstract The number of people diagnosed with advanced stages kidney disease have been rising every year. Early detection and constant monitoring are the only minimally invasive means to prevent severe damage or failure. We propose a cost-effective machine learning-based testing system that can facilitate inexpensive yet accurate health checks. Our proposed framework, which was developed into an iPhone application, uses camera-based bio-sensor state-of-the-art classical learning deep...