About
UTS Tech Festival (19 – 30June, 2023) is a two weeks festival of events, showcases, hackathons, workshops, masterclasses, seminars, competitions, industry-student engagement and much more! Our goal is to bring together students, academics and industry to foster learning, inspiration, share ideas and promote innovation.
Come along to the UTS AI Showcase which is part of the UTS Tech Festival 2023, to discover Artificial Intelligence and Deep Learning!
This the second iteration of the UTS AI Showcase that showcases Artificial Intelligence (AI) and Deep Learning projects developed by undergraduate, postgraduate, and higher degree research (HDR) students from the Faculty of Engineering and IT (FEIT).
Venue:
UTS Faculty of Engineering and Information Technology
Broadway, Ultimo NSW 2007, Australia
Date and Time:
Wednesday, June 28, 2023, 2.00pm – 7.00pm
Showcase Winner
Best Project: Innovation/Research
Team: PronAI
Team Members: Malena DIAZ RIO, Benjamin LAWRIE, Rodrigo DE MIGUEL SATUE
Best Project: Usecase/Application
Team: InkWell AI
Team Members: Abdulaziz EL SABBAGH, Ihtsham SHAFIQ, Lachlan CROLL
People Choice Award
Team: Plastic Bottle Detection
Team Members: Yashkumar Patel, Cormac Hegarty, Charith Reddi Mekala N
2nd Best Project:
Team: Poisonous Spider and Snake Detection
Team Members: Shawn WOO, Fanny PUTRI
SAS Industry Choice Award
Team: PronAI
Team Members: Malena DIAZ RIO, Benjamin LAWRIE, Rodrigo DE MIGUEL SATUE
Special Mention by Judges
DeepERV
Team Members:
Nikhil Deo
Special Mention by Judges
SeagulEye
Team Members:
Gerardo Bohorquez Restrepo
Javier Chapto
Special Mention by Judges
Online Age Gate
Team Members:
Special Mention by Judges
Tetris with gesture detection
Team Members:
Justin Qi,
Nicholas Madew,
Joel Morrison
Photo Gallery
Program
Participating Teams
Where's Waldo
An innovative image classification app designed to amplify the VIVID Sydney festival experience by spotlighting the artists behind the art installations. The app's primary motivation is to provide a platform for these talented artists to be recognised and appreciated for their remarkable contributions!
Room : CB11.04.400
Pod : 1
PronAI
Pronation refers to the way our feet hit the ground. Traditionally, the pronation analysis is done by professionals. Our project aims to automate this analysis and deploy it to allow public access. In order to automate this analysis deep learning is used. Instance segmentation & mathematical methods are applied to classify pronation.
Room : CB11.04.400
Pod : 2
EyeCandy
This project proposes a Self-Checkout System for Cafeterias that leverages deep learning and OpenCV to automate the process of ordering, payment, and checkout in a cafeteria. The system uses object detection and recognition algorithms based on deep learning to identify the items selected by the customers.
Room : CB11.04.400
Pod : 3
Accessorise
Online Shopping provides plethora of options but uncertainty with regards to which jewellery to buy and which would suite the buyer. Introducing Accessorise: an online jewellery shopping website which allows users to try on a necklace prior buying them. User can try on necklaces and decide which one to buy.
Room : CB11.04.400
Pod : 4
Efficient Waiters
Our project is designed to improve performance efficiency at restaurants, bars, pubs etc. Specifically, our project detects when a customer has an empty glass on their table and alerts staff, thus allowing them to refill or be aware of their empty glass quickly through the power of deep learning.
Room : CB11.04.400
Pod : 5
NaviCane
This project revolves around developing a wearable smart glass system that leverages deep learning based object detection and audio navigation technology to aid visually impaired individuals in safely & efficiently navigating their surroundings, enhancing their mobility, & overall quality of life.
Room : CB11.04.400
Pod : 6
InkWell AI
Inkwell AI is a project that aims to provide a unique and innovative solution for the tattoo industry. InkWell provides a fun and interactive way for people to experiment with different tattoo designs, a valuable tool for tattoo artists to promote their work online and streamline the tattooing process.
Room : CB11.04.400
Pod : 7
Audio Insights
This solution detects musical instruments from songs or audio files using a variation of Convolutional Neural Network (CNN). The objective of the model is to analyse the spectrogram of the audio files to determine which instruments are present within the audio.
Room : CB11.04.400
Pod : 8
Student Distraction Behaviour Detection
The application aims to address the issue of student distraction in the classroom by utilizing object detection techniques. The application serves as a valuable resource for teachers, helping them monitor and minimize distractions during class sessions.
Room : CB11.04.400
Pod : 9
Smart Travel
Situational awareness is essential for autonomous navigation of drones for various use-cases such as delivery of medical supplies, food, etc. Smart Travel provides situation awareness to the drones, enabling autonomous navigation of drones/UAVs, by avoiding collision with other drones and obstacles.
Room : CB11.04.400
Pod : 10
Style Detector
Style detector is a smart fashion recommendation system that provides users with clothing recommendations based on an input image. Using this system, any user who wants to look stylish, but does not necessarily have the time to research the latest fashion trends, can easily find an entire outfit consisting of a top, bottom and jacket.
Room : CB11.04.400
Pod : 11
DeepERV
Every second counts when responding to emergency situations. Traffic can directly affect the time it takes for help to arrive. Our solution aims to inform when an active Emergency Response Vehicle (ERV) is approaching so drivers have more time to move out the way.
Room : CB11.04.400
Pod : 12
SceneSpec
SceneSpec revolutionizes local shopping by leveraging Computer Vision and Deep Learning. It detects store boards live, fetches real-time reviews based on location, and provides key store information such as contact details, working hours etc. This innovative mobile app promotes informed shopping decisions while enhancing the visibility of local businesses.
Room : CB11.04.400
Pod : 13
SeagulEye
This project consisted of performing object detection (people) on beach scenarios, looking to help beachgoers make informed decisions on the best times to visit the beach based on their preferences, The final result consists of a web application that gives users access to videos of beaches or their camera and the possibility to get information on how crowded the beach is.
Room : CB11.04.400
Pod : 14
FireFly
An innovative AI-powered solution designed primarily to detect bush fire. By harnessing the power of drones equipped with cutting-edge fire detection capabilities, Firefly scans volatile environments, minimising the risk of human error and significantly enhancing fire detection accuracy. This technology is showcased through a user-friendly web application.
Room : CB11.04.400
Pod : 15
Poisonous Spider and Snakes Detector
Our application can detect poisonous spiders and snakes which can guide individuals who encounter them either inside their houses or outside in the backyard and wilderness. It can recognize snakes and spiders, and provide first aid guides for snakes and spider bites.
Room : CB11.04.301
Pod : 1
Online Age Gate
As the internet rapidly expands, so do the potential online risks such as gambling, alcohol purchase and online dating. Online Age Gate is a deep learning solution, designed to help safeguard both websites and users alike by 'Gating' access to risky online sites via age detection. Through computer vision, users can be classified as above or below 18, with the solution intelligently granting or denying access.
Room : CB11.04.301
Pod : 2
Anomaly Detection in Surveillance Videos
CCTV is commonly used for surveillance in private and public sectors. However, the conventional approach of manually reviewing and analysing CCTV footage might take time and effort. This project will uses a web-based application to be able to detect if a surveillance video recording contains any anomalies.
Room : CB11.04.301
Pod : 3
Tetris with gesture detection
To redefine how the household classical game, Tetris, is played. Enabling controls via intuitive head movements. This implementation leverages gesture detection through a deep learning model trained on facial landmarks. Our work underscores the potential of gesture-based interfaces in gaming, & accessible user experiences.
Room : CB11.04.301
Pod : 4
Not A Robot
In the current digital era, news and information have the ability to spread quickly online to entire populations of people. 'Not-A-Robot' aims to contribute to broader efforts in combating the risks of deepfakes by raising awareness, providing accessible tools, and promoting a safer and more informed online environment.
Room : CB11.04.301
Pod : 5
Pre-Cooked meal detection
Self-service checkout systems usually require a high level of customer involvement. This automatic cooked food checkout system improves the customer shopping experience at food court. We use image recognition techniques to detect the quantity and type of food placed on tray.
Room : CB11.04.301
Pod : 6
Road crack detection
The goal of our team's project is to develop a system for automatic road crack detection. The system program supports the localization, identification and marking of road crack types in pictures, videos and live cameras. So, it can be useful in tasks such as pavement maintenance or road safety assessment in cities.
Room : CB11.04.301
Pod : 7
Locating and Identifying Ships from Satellite Images
Ships play a crucial role in transportation, trade, maritime security and many other areas. However, manually monitoring vast stretches of water bodies is a labor-intensive and time-consuming task. Using CNN-based object detection, an automated solution was developed, with vast applications potential.
Room : CB11.04.301
Pod : 8
Plastic Waste Detection
The impact of climate change is getting more and more perceptible with each passing day. Plastic contamination is a significant factor in the issue. Research shows, only around 9% of the 500 billion plastic bottles we use year are recycled. This project investigates the identification of plastic bottles/wastes using computer vision. A smart mobile app is presented in this project.
Room : CB11.04.301
Pod : 9
Pedestrian Density Estimation
This project proposes a web-based solution for automatic estimation of pedestrian density in a specific area. The system uses deep learning model to detect and count individuals in images or live video streams. The density estimation is divided into 3 levels, which are distinguished by different colors. This helps to enhance public safety and optimize resource allocation in crowded environments.
Room : CB11.04.301
Pod : 10
Gardeners Friend
The amateur gardener is often not an expert at identifying weeds in their garden and have limited knowledge on appropriate control measures. This can lead to expenditure on ineffective control measures and unnecessarily expose gardeners to hazardous chemicals. Using object detection techniques, models were trained on images of weeds dataset and can be displayed with our web app.
Room : CB11.04.300
Pod : 1
Fake face detector
The evolution of GANs has enabled the creation of fake profiles with faces which are incredibly hard to distinguish from real faces. When faced with such an issue, our group decided to create a method using CNNs to identify fake faces from pictures given and provide a response to the user, identifying if the image is fake or not and the confidence level of the model.
Room : CB11.04.300
Pod : 2
AutoStock
AutoStock aims to improve stock management efficiency through object detection. The system can prevent resource waste and optimise inventory control by automating stock monitoring and tracking using computer vision algorithms. We solve and improve the efficiency of the stock management system through object detection aligns with the goal of preventing unnecessary waste of resources in stock management.
Room : CB11.04.300
Pod : 3
Coral Bleaching Detector
Our project explores the application of Deep Learning-based object detection technique in detecting bleached coral. Inspired by the need to monitor and conserve coral reefs, we leverage the speed and accuracy of Yolov7 to identify and track bleached corals in real-time. By utilizing this advanced detection method, we aim to improve the efficiency and effectiveness of coral bleaching monitoring efforts.
Room : CB11.04.300
Pod : 4
Nail Health
Unveil the secrets of your nails with Nail Health: Decoding diseases at a glance! Nail Health is an advanced solution that harnesses image processing and deep learning techniques to accurately detect and classify various dermatological nail problems. By analyzing the visual cues from an individual's fingernails, Nail Health provides valuable insights into potential diseases. Experience a revolutionary approach to nail care with Nail Health.
Room : CB11.04.300
Pod : 5
Drive Guardian
Driver drowsiness has been a recurring issue around the world. It causes incidents to occur and endangers the lives of many people around the streets. Our team wants to create an environment that prevents drivers from driving when they are drowsy by warning them. To reduce the chance of having a drowsy driver using their car on the road, our team had an idea to create a detection system that detects when a driver looks drowsy.
Room : CB11.04.203
Pod : 1
DeepClean
Underwater waste caused by humans remains an ongoing issue for the marine ecosystem. The introduction of Autonomous Underwater Vehicles (AUVs) has provided the opportunity to leverage the integration of other technologies into underwater exploration. Our project, DeepClean, is a system that is aimed to be integrated with AUVs to detect underwater plastic wastes and support ocean cleaning efforts.
Room : CB11.04.203
Pod : 4
Projects demos from the
Data Science Institute (DSI) &
Data Science and AI Society
Digital twins platform for Smart Cities
Our "Digital Twin platform for smart cities" was born from the desire to offer various stakeholders the opportunity to tackle pressing urban challenges by fusing multiple data sources of different types, from geo-spatial mapping, and 3D BIM models to real-time data streams around people’s movement. With a focus on sustainability, the platform offers solutions for improving public transportation, building management, air quality, and smart parking.
Room : CB11.04.203
Pod : 2
Intelligent Scanner
Automatic detection of prohibited items in real-time from X-Ray screening images of baggage. Intelligent Scanner, an AI-based X-ray scans monitor.
Room : CB11.04.203
Pod : 3
Meet Our Judges
Ian Edwards
SAS
Harshini Jayakumar
SAS
A/Prof. Hai Yan Lu
University of Technology Sydney
Dr. YK Wang
University of Technology Sydney
Dr. Muhammad Saqib
CSIRO/Data61
Meet Our Industry Speakers
Ian Edwards
As Head of SAS’s Academic Outreach Program, Ian is responsible for leading the program the purpose of which, is to "Connect Corporate Australia and Higher Education to power Analytical Capability and Innovation".
Ray Greenwood
Currently the manager of the Customer Advisory team for Government at SAS Australia & New Zealand. Prior to this, Ray has spent two decades in the world of advanced analytics as both a user and a vendor of analytics platforms across a diverse range of industries including financial services, retail, government, telecommunications and more. He has found that if there is one consistent theme across the years, countries, businesses and industries it is that there is always business value to be found in innovation driven by analytics.
Organizing Committee
A/Prof. Nabin Sharma
Organizing Chair
Gitarth R Vaishnav
Event Lead
Jason Do
Coodinator, Student Engagement
Sarah Rodriguez
Student Engagement Officer
Vishesh Sompura
Robotics Lead
Verren Lo
Event Assistant
Sohitha Muthyala
Photography Assistant
Supported By
