CSE 2023 Fall Departmental Demo Day
We’re thrilled to invite you to the tenth bi-annual Comp. Sci. & Eng. Fall Demo Day. Student groups from several CSE capstone classes will be presenting the culmination of 3-months of effort, hard work, (metaphorical) blood, sweat (well… caffeine really), and tears (see above).
Where: Davis Hall; 1st and 2nd Floor Atrium
When: Monday December 11th, 2023
Schedule
2:00 PM -
Event opens
2:05 PM -
Opening remarks
2:10 PM -
Judging starts
4:00 PM -
Judges will get together to review results and pick winners
4:30 PM -
Winners announced and videos played for the winning teams
5:00 PM -
Event closes
Winners
Acknowledgements
Sponsors
Judges
- Chris Miller
- Arnav Matta
- Anarghya Das
- Craig Ross
- Andrew Olewnik
- Neel Tripathi
- Samantha Seymour
- John Tantillo
- Victoria Faltisco
- Amir Nassereldine
- Ethan Blanton
- Greg Bell
- Jim Brandt
- Jaric Zola
- Haonan Lu
- Roshan Ayyalasomayajula
Presented Projects
This year’s participating classes and projects include:
(CSE 611: Master's Capstone Project)
- ACV Auctions
-
- Harshitha Reddy Jaggavarapu
- Shiva Karthik Reddy Bathula
- Yashwanth Krishna Porandla
- Sufianuddin Mohammed
Cutting-edge mobile application.Say goodbye to the hassle of manual car appraisals—our app takes the guesswork out of determining your car's worth by utilizing live car video recording. Whether you're looking to sell, trade, or simply curious about your car's current value, our app is your go-to companion.
- Carbon Tracker Data wrangling
-
Updating the old databases with the latest census information using the API calls. Also, There is a My PM2.5 webpage that shows the particulate emissions due to transportation, heating and electricity which causes health impacts. So, this project gives a perspective on carbon emissions and particulate emissions.
- Exhibit QR Code
-
Introducing our cutting-edge "Exhibit QR Code" Management System, a powerful and intuitive solution designed to streamline the organization and accessibility of exhibit data. This comprehensive web-based interface empowers users to effortlessly manage, update, and showcase a diverse range of exhibits, each assigned a unique QR code for instant access.
- Fresumes
-
Fresumes: Revolutionizing recruitment and job hunting! Explore our free resume website offering a robust resume builder and parser for candidates to create compelling profiles. For recruiters, our platform facilitates seamless candidate searches, mass emails, and messaging—making Fresumes the ultimate hub for efficient hiring and standout career building. Streamline your hiring process with Fresumes, where talent meets opportunity seamlessly.
- Health 360
-
Unveiling a revolutionary platform catered to the global medical community, our innovative website is a comprehensive hub designed for medical professionals, researchers, and practitioners engaged in clinical trials. Users can seamlessly create and manage trials, personalize them with distinctive names and detailed descriptions, and integrate custom questionnaires effortlessly. The platform's versatility extends to the creation of user groups within trials, facilitating targeted engagement through tailored notifications. Practitioners can efficiently notify specific user groups for questionnaire completion, fostering a collaborative and engaged research environment. The system's efficiency shines through in its user-friendly tabular format for response analysis, with the added convenience of downloading responses in Excel format for in-depth studies. Complementing this, our user-friendly mobile app ensures accessibility for all, allowing easy sign-up, log-in, and participation in trials curated by researchers. This intuitive interface promises a holistic and engaging experience for both researchers and participants, marking a significant advancement in the landscape of clinical research.
- Invoicing system with Jira Integration
-
- Sai Prasanth Kethavarapu
- Sai Pavan Kumar Yadlapalli
- Lakshmi Naga Khyathi Chittineni
- Sai Sridhar Reddy Palli
This project introduces an efficient invoicing system integrated with Jira, aimed at streamlining billing processes for digital service businesses. Key features include real-time synchronization with Jira to track service tickets, automated invoice generation, and a user-friendly interface. The system enhances workflow efficiency, ensures billing accuracy, and includes an inventory management module. Developed using React, Node.js, MySQL, and hosted on AWS, it offers a reliable, scalable solution for improved client satisfaction and operational effectiveness. Future updates may incorporate AI for advanced inventory management.
- Offset CRBN
-
Offset CRBN is a comprehensive web platform designed for environmental awareness, allowing users to easily calculate and manage their carbon footprint. The user-friendly carbon calculator, customizable questionnaire, and administrative dashboard empower users to make informed decisions. Admins have control over questions, prices, and formulas, ensuring accuracy and adaptability. The platform also features robust customer support, emission data visualization, and aims to contribute to a more sustainable future by raising awareness and encouraging eco-friendly choices.
- Online Dictionary
-
Online Dictionary is an web application designed for word lookup, including definitions, etymology, synonyms, antonyms, and pronunciations. The primary aim of this product is to assist various user groups, including students, academicians, and professionals, by offering free access to dictionary services. This initiative benefits educational institutions by providing ad-free and spam-free access to dictionary. Our team is committed to enhancing the current system by introducing features such as user login for personalized experiences, word categorization, lookup history, favorites, and language-based suggestions.
- Online Event Scheduler
-
The Online Event Scheduler is a web-based application designed to assist individuals in efficiently scheduling, verifying availability, and organizing both individual and group meetings for University at Buffalo (UB) students and faculty members. This platform eliminates the need for manual availability checks and streamlines the process of identifying a suitable common meeting time. The application prioritizes security measures and offers a user-friendly interface, ultimately saving users valuable time.
- Photo Editor
-
The Photo Editor Web Application is a versatile and user-friendly tool designed for enhancing and manipulating digital images. Offering a comprehensive set of features, this web application bridges the gap between basic photo editing tools and advanced image editing software. The project targets a diverse audience, including individuals, hobbyists, amateur photographers, professionals, students, and businesses of all sizes. Key Features: Advanced Passport Creation: A unique feature allows users to enable/disable the 'apply white background' offering the choice between using high-end tools or minimalistic features when generating passport images. Offers a wide range of country selection and cropping based on the specific country requirements. Editing Tools: Users can enjoy a variety of tools such as cropping, resizing, noise removal, collage creation, mosaic application, and brightness/contrast adjustment. Image Format Support: The application supports various image formats, including JPEG and PNG, ensuring compatibility with commonly used file types. Compression Features: Users can compress both images and videos, optimizing file sizes for efficient storage and sharing. User-Friendly Interface: The software boasts a clean and intuitive interface, providing a seamless experience for users to navigate and apply editing tools with ease. Output Options: Users have the flexibility to save edited images in different file formats.
- Sports Perfornance Pro
-
Ultimate toolkit for elevating athletic excellence! Unleash the full potential of your team with the cutting-edge application designed for coaches and athletes. Effortlessly create, manage, and track customized workout plans, while seamlessly monitoring and optimizing athlete performances. Elevate your training game, maximize results, and conquer the competition with Sports Performance Pro – where precision meets peak performance.
- Team TukTuk
-
TukTuk is a ridesharing application developed by a team in Austin, TX. The team at Buffalo is working with the Austin team to develop a revamped application suitable for Buffalo. The application is meant to provide an easier and safer mode of transportation for UB students to travel from one hotspot to another within or outside the campus. The app is also meant to act as a source of revenue for UB students by offering rides to other students while travelling around Buffalo. The application is being developed ni React Native, Node js and AWS services for deployment and storage.
- Web Tool for Words and Phonemes
-
The project aims to create a word database with metadata and phoneme information, offering an API for querying this database and a user-friendly web interface. It will include features such as scoring words based on popularity, complexity, readability, and sound, storing metadata like categories and semantics. displaying words based on specific phonetic queries, and presenting statistics on phonemes. The tool is used to assist speech-language pathologists and can be used to help children with pronunciation problems, teachers for educational purposes, and general users looking to query and explore word-related data
- Youro
-
Youro is your comprehensive solution for optimal urologic health, offering innovative features to simplify and elevate your healthcare experience. Track your symptoms with our exclusive calculators, ensuring personalized care and treatment. Schedule same- or next-day appointments with urology-trained providers, fostering continuous communication through our app. Access up-to-date educational materials related to your diagnosis, empowering you with knowledge. Benefit from individualized care plans designed for your specific needs, guiding you towards improved symptoms. With Youro's digital-first approach and 24/7 availability of urology-trained providers, achieving peak urologic health is simple, enjoyable, and tailored to you.
- CMS Mavens
-
Website, CMS system, admin portal and member portal for the AI for Exceptional Education organization headed by UB
- museum Interactive exprience
-
The objective of this project is to develop an application for museum patrons to participate in a scavenger hunt application. The patrons would be given a challenging set of scavenger hunt questions, they would have to find the exhibit that satisfies the answer to the question and scan a near field communications (NFC) chip with their smartphone to get credit for answering the question. The application would time the execution of the scavenger hunt and patrons will see their ranking at the end of the scavenger hunt. A companion objective is to develop an administrative application to load and maintain the scavenger hunt information, narrative, photos, etc. for consumption by the patron application. The senior design class will be responsible for defining the project requirements, creating a comprehensive plan, and overseeing its development and deployment
(CSE 302)
- Android And Java Malware Detection
-
This project aims to create a toolchain based around a compiler and machine learning classifier to identify Android and Java malware. Students will work on modifying an existing compiler (Soot) to analyze Java and Android programs. The project will involved writing static and dynamic analyses to identify features that can be used by a classifier to make a determination if a given program is malicious or not. The project will begin with students learning about and writing their own intra-procedural information flow analysis. Students will then extend their intra-procedural information flow analysis to be a whole program inter-procedural information flow analysis. This analysis will identify the portion of the computation that deals with sensitive and personal data. Students will then write an analysis over this portion of the code to extract features to be used for classification. The analyses can be extended to handle more complex Java and Android features like reflection and Android IPC mechanisms. There is also room for exploring different machine learning technologies based on performance and accuracy measures.
- Manufacturing metaverse
-
We've created a 3D manufacturing metaverse housing an autonomous robotic arm equipped with the ability to detect plastic from a pile of different materials. Leveraging Unity and Blender, we've brought this innovative concept to life. Our 3D manufacturing environment serves as a valuable testing ground for manufacturers, allowing them to experiment and refine their processes with the help of our autonomous robotic technology.
- Teamwork Tool
-
Our platform empowers instructors with an intuitive UI/UX, facilitating seamless course, survey, and rubric management.
- Vizier
-
Vizier is an advanced, reproducibility-focused notebook environment that supports multiple languages like Scala, Python, and SQL. It uniquely features a spreadsheet mode for direct dataframe editing, templated commands for streamlined data tasks, and an automatic cell update mechanism to maintain data freshness. Additionally, Vizier offers comprehensive history tracking for revisiting or branching analyses and includes a Dataset Stabilizer to monitor and alert about significant data changes, making it a highly adaptable and user-friendly tool for complex data workflows. The Vizier team is rebuilding their computational notebook Vizier, and wants to take this opportunity to develop a simpler, but more powerful data visualization component. At a really high level, this component takes in a dataset and spits out a chart based on a set of user-provided instructions. This semester, the Vizier team worked to redesign the UI/UX of the editor to allow an easier user experience with the same performance as the previous editor. We introduced a Bar Chart Command where you can see/use our editor. In our editor, we have introduced easier way to add elements, choose filters, change styling, etc.
(CSE 410: Game Design)
- Bounce Tower
-
An arcade platformer game made with Unreal Engine 5
- Echoes of the Fog: A Skinwalker's Curse
-
A horror game made with Unreal Engine 5. This game showcases mechanics made with Unreal Engine's Blueprint system, and assets from the Unreal Engine market store. The goal of this game is to collect pages around the map and make it back to the original spawn point to click the red button and win the game.
- Eyes Wide Open
-
A horror-themed video game where the player must collect their stopwatches to escape statue like creatures that are hunting you. (Game description/tag in other details)
- Frogger's Lane Lunacy
-
Frog Driving Simulator
- Into Despair
-
Into Despair is a first person horror game where you wake up after a night with your friends to find them gone. Venture into the forest and find out its secrets.
- Restaurant'botage
-
Restaurant'botage is an FPS-style cooking simulator 3D game. You work at a diner that has customers barging in, demanding food. It is your job to serve them as fast and as accurately as possible before they take over!
- Salvation Sky
-
A survival video game, taking place among the sky. Traverse and survive on Sky Islands and repair your ship to make your way home.
- Stellar Odyssey
-
Stellar Odyssey is a physics based puzzle game akin to the Portal or Half-Life family of games. You must solve puzzles across many levels to escape an abandoned space station, using three different abilities: anti-gravity, stasis, and rewinding.
- Student Simulator
-
Student Simulator is a third person stroy game that follows the story of Mark, a fourth year transfer student to a university. The game involves the main objective of graduating on time and avoiding choices that lead to him dropping out. He is your typical, average computer science major student but throughout the game, he wil be involved in situations like choosing who becomes friends with, choosing the right words to say, amongst other situations that enable players to experience a game that develops depending on the decisions that are made.
- Terminal 497
-
A horror space videogame where you must escape the alien infested space station.
- Tower DeFense
-
A game that combines tower defense and first-person shooter. Players are armed with both construction blueprints and an arsenal of weapons. As monsters assail your shelter, players will decide how they protecting their shelter: will they bolster their defenses by erecting towers, or take a direct approach, confronting invaders face-to-face.
(CSE 442: Software Engineering)
- BlueBrilliant
-
A Chess engine website built in javascript
- Cat Crowd
-
This project is a simulation app in which users can play minigames and customize their own cat cafe.
- Chronline
-
Chronline is a media logging app which integrates tracking across books, games, and movies. The main purpose of our app is to use these patterns to calculate your chronically online score, and to improve the quality of your life by reminding you when it’s time to go outside or socialize. You can also post, like and comment on other people or your friends’ reviews to find like-minded individuals and get better at expressing yourself. We take your security seriously, and have features such as blocking and making your account private to control the visibility of your information. There’s something in Chronline for everyone!
- Grouped / Len[cs]
-
A collaborative planner for your everyday life making scheduling and communication seamless.
- Helping Hands
-
Helping Hands is a web application designed to revolutionize the volunteering landscape. It serves as a bridge between volunteers seeking meaningful opportunities and organizations requiring assistance. This platform uniquely caters to the needs of busy individuals like students and professionals, as well as non-profit organizations striving for effective volunteer management.
- Hungry All The Time
-
The system enables customers to place food orders, maintain their personal accounts, complete checkouts, utilize credit and promotional codes, and provide reviews and ratings for their orders. It also provides merchants with the ability to update their products, access order statistics, and review customer feedback. Additionally, administrators have the capability to manage user accounts, order data, and review information.
- Shrubbery Quest by Nefariouz Gamez
-
Shrubbery Quest is a 2D skill-based platformer game where you fight your way through 20 unique and challenging levels to reach the golden shears -- chopping down shrubs and (successfully or unsuccessfully) avoiding untimely deaths along the way.
- Nutrient Fanclub
-
Our project is a web application that allows students going to UB to leave reviews for restaurants around or in the campus area.
- Gig App/ Fugue State
-
Our app allows the user to create their own account so they can create music from scratch, practice instruments and songs and look up songs.
(Research Under Dr. Ratha)
- Chat-based Secure Disease classification based on symptoms using FHE and Reinforcement Learning
-
In this work, we develop a secure potential disease classification based on the symptoms provided by the user. We developed a customized fully connected layer and utilized a ReLU approximation activation function that could operate on CKKS-encrypted data.
(CSE 587)
- Crimson Eye: A Data-driven Approach to Crime Analysis
-
Crime is a complex issue that affects every part of the globe and poses a major threat to community safety and well-being. It is essential to understand and tackle crime in a way that safeguards people and nurtures a sense of peace in our society. In the midst of all the challenges the police forces grapple with, our project can act as a guiding tool to deploy resources more effectively and prevent crime. In other words, predictive policing is a method that makes use of past social, demographic, and economic crime data to spot trends and estimate where and when a crime is most likely to occur. Insights from our application, serve as the foundation for creating effective strategies to prevent and intervene in criminal activities. We provide a data-driven method of analyzing crime that can be proactively used to reduce the crime rate and foster a safe community.
- Global Climate Change Analysis of the Major Cities over the Years
-
Our project focuses on to present the change in the climate over the years in the major cities of world. This change is majorly determined by the global temperature over the years.
- Mental Health Challenges in the Corporate World
-
Our project involves utilizing a dataset comprising surveys from individuals experiencing mental health issues. The primary objective is to investigate whether individuals, despite receiving resources from their respective companies, undergo treatment and successfully recover from their mental health challenges.
- Radix Nexus
-
Hybrid AI-Driven Framework: Optimizing Initial User Engagement in Recommender Systems
- US Accident Prediction
-
Road accidents pose a significant threat to public safety, and understanding the underlying patterns and factors is crucial for effective prevention and mitigation strategies. This project aims to analyze a countrywide traffic accident dataset covering 49 states in the United States, comprising data from diverse sources such as transportation departments, law enforcement, and traffic sensors. The primary objective is to uncover insights into the causes, frequency, and impact of accidents, ultimately contributing to the development of informed policies and interventions for enhancing road safety.
(Research Project under Prof. Nalini)
(Research )
- Leveraging FHE for Enhancing Soft-Biometric Privacy
-
Face recognition, a cornerstone of early deep learning methodologies, has become a pivotal technology in AI systems, raising concerns regarding privacy and security in widespread applications. Face embeddings, though essential for identity verification, have been shown to be susceptible to leaks, enabling unauthorized access and unintended information extraction. This research investigates existing template protection techniques and their effectiveness in preventing biometric information leaks. Our findings reveal vulnerabilities in many existing techniques, emphasizing the need for enhanced privacy safeguards. To address this, we propose a novel dual-layer protection method for face embeddings, combining established protection schemes with an additional encryption layer to fortify privacy. The use of Fully Homomorphic Encryption (FHE) enables secure analytics on encrypted templates. Our approach successfully circumvents the shortcomings of individual protection techniques, providing an elevated level of privacy and security. We demonstrate the effectiveness of our method through face verification, age, gender and ethnicity prediction tasks, achieving comparable performance to unprotected templates. Leveraging FHE alongside conventional protection methods ensures irreversibility and unlinkability, preserving accuracy while eliminating the risk of biometric information leaks. The project presents comprehensive results supporting the efficacy of our proposed approach in advancing the privacy and security landscape of face recognition systems.
(CSE302)
- PWNIoT
-
We helped prepare the launch of Dr. Ziming Zhao's cybersecurity education platform PWNIoT.
- ML model recommendation and benchmark systems
-
The purpose of this project is to be able to integrate Machine Learning models onto the MLModelScope platform; A hardware/software agnostic allowing novice users to run and benchmark Machine Learning models. For our demo, we are featuring the testing and running of two models.
(Research Project Under Prof. Nalini Ratha.)
- Secure Sleep Apnea Detection using FHE
-
We work on a privacy preserving sleep apnea detection method from ECG signals using Fully Homomorphic Encryption (FHE) and Deep Learning. FHE enables deep learning analysis on encrypted ECG signals thereby guaranteeing security and privacy.
(CSE 666)
- Secure Vascular Biometric Recognition
-
In this paper, we propose an end-to-end vascular feature recognition system built with the security and privacy of the template in mind. Our solution uses Resnet50 architecture coupled with biohashing to securely store and manage the biometric data of individuals. The system makes use of the images captured using a near-infrared light scanner. This image is processed to generate a biohash, a unique digital representation of the user’s vein pattern. The system also employs a matching algorithm that compares the user’s biohash with the stored biohash to verify the user’s identity. The core model achieved an accuracy of 95.58% compared to the state-of-the-art accuracy of 97.66%. However, our proposed solution achieves an EER rate of 0 after biohashing whereas the state-of-the-art model has an EER rate of 2.03, indicating 100% efficiency of biohashing.
(Independent Study/ Project)
(Research)
- Fully Homomorphic Encryption Operators for Score and Decision Fusion in Biometric Identification
-
This research introduces an approach to overcoming the limitations of unimodal biometric systems through the development of a secure multimodal biometric system. The core of this system lies in its encrypted score/decision-level fusion, which integrates various biometric traits such as face, fingerprint, and iris. Utilizing the power of non-interactive homomorphic encryption, this system significantly enhances the accuracy, security, and privacy aspects, making it particularly suitable for cloud-based biometric services.