This post has been republished via RSS; it originally appeared at: Student Developer Blog articles.
First published on MSDN on Jun 08, 2018Now in its 16 th year, Microsoft's Imagine Cup has a mission to empower student developers to bring their technological innovations to life. Competing students create technology solutions with the potential to change the world.
We are excited to announce the 49 finalist teams who will be coming to Seattle for the 2018 World Finals this July. Students will compete for the chance to win up to $100,000 and a mentoring session with Microsoft CEO Satya Nadella. In addition, three special $15,000 prizes were also introduced this year to challenge students to bring forward their ideas in Artificial Intelligence, Big Data, and Mixed Reality. A trio of Finalist judges, all with extensive experience in technology and business, will pick the next Imagine Cup Champions.
Students this year have truly inspired us with their groundbreaking ideas for the future of the tech world. From a machine learning algorithm that detects the presence of wildfires, to a HoloLens app that improves dyslexic children’s reading skills, and an AR/VR platform that allows engineers to be holographically teleported into a workspace, this year’s competition will be tough. Follow along with their journeys real-time to see who will win.
Here are the World Finalist teams traveling to Seattle in July, bringing their inspiring projects to a global stage:
Lexa |
Argentina Instituto Tecnologico de Buenos Aires |
Lexa is a platform for giving transparency, eliminating fraud, and bringing the digital era to the medical industry. Through the use of open blockchain technology and machine learning algorithms, the goal is to increase patient trust in the healthcare system. |
|
Belarus American University of Armenia State Institution of Education, Secondary School №11 the town of Slutsk |
Our project uses a neural network to determine psychogeometric characteristics of a person from photographs, which allows more effective communication according to their individual characteristics, with a focus on business and sales settings. |
|
Brazil Centro Europeu |
ADAM Robo is a visual acuity screening software used to test and prevent common vision problems. ADAM uses bot conversation guiding an individual along anamnesis and vision tests which can identify possible refractive errors such as myopia (nearsightedness), hyperopia (farsightedness), presbyopia (loss of near vision with age), and astigmatism as well as Daltonism (color blindness). |
smartARM |
Canada University of Ontario Institute of Technology |
smartARM is a robotic hand prosthetic, created using Microsoft Azure Computer Vision, Machine Learning and Cloud Storage. The robotic hand uses a camera embedded in its palm to recognize objects and calculate the most appropriate grip for an object. Using Machine Learning, the more the model is used the more accurate it becomes. |
Muma |
Canada University of Toronto |
Muma created a music matching service that can help identify music usage online to capture revenue for its customers. The project combines multi-layered convolution neural networks and data analysis or peripheral information (such as comments on a YouTube video) to produce accurate lyrics, and also to make a wide range of content accessible to the hearing impaired. |
Hachy |
Canada University of Waterloo |
Hachy aims to provide an easy way for farmers to identify the development status of an egg by enabling autonomous egg development identification and tracking using Machine Learning. |
ShowNM |
China Tsinghua University |
ShowNM helps to overcome the difficulties of quantitative diagnosis of Parkinson’s disease. Combining deep learning with signal processing, this project utilizes the convolutional neural network to extract features from Parkinson’s disease patients’ surface electromyography. It then can grade their states according to a nonlinear classifier. With this method and its small, light instrument, doctors will be able to give the right therapy to patients. |
Peppa Team |
China Sichuan University |
Peppa Team created a toy car that can help diagnose ADHD and train kids’ ability to focus. The app can provide the results of early examination, real-time detection, and adjuvant therapy. It is not only a wonderful combination of brain-computer interface and deep learning, but also a breakthrough in the application of Recreation Therapy, achieving painless and convenient diagnosis and treatment. |
BeachSafer |
Egypt Cairo University |
BeachSafer aims to make swimming safer using a smart integrated IoT solution. It provides a real-time monitoring and supervision dashboard on a swimmers’ state and location through a lifeguard mobile app and smart wearable devices for swimmers. Biometrics are used to detect drowning risk and initiate an automated rescue procedure. |
Theatrall |
France Ecole Polytechnique |
Theatrall aims to make theaters accessible for everyone through simple software available on smartglasses, smartphones or tablets, which will display the production subtitles in the language chosen by the user. |
Soul Sailor |
Germany Technical University Munich |
Soul Sailor is a platform that provides psychological care for refugees and asylum seekers, including a chatbot to help individuals process their experiences while seeking refuge. |
NASC |
Germany Karlsruhe Institute of Technology |
NASC is a web app that searches for news articles on the web and attempts to evaluate sentiments in the articles. The goal is to encourage people to read beyond the first search result for important stories, and to look at multiple articles that express different views. |
Pavo |
Germany
Rhine-Waal University of Applied Sciences |
Pavo Vision makes websites more accessible to visually impaired users. Utilizing AI, images, and other tools, non-readable website content will be analyzed and made available to visually impaired people. Mistakes in analysis can be fixed and the algorithm will get smarter over time by utilizing crowd-sourcing to generate more accurate descriptions. |
StudySmarter |
Germany
Technical University of Munich |
StudySmarter utilizes Machine Learning to provide students with a platform to make learning more efficient and motivating. Students are connected with each other, and algorithms automate and create learning materials and provide feedback on individual progress. |
iCry2Talk |
Greece
Aristotle University of Thessaloniki |
iCry2Talk proposes a low-cost and non-invasive intelligent interface between the infant and the parent that translates in real time the baby’s cry and associates it with a specific physiological and psychological state, depicting the result in a text, image and voice message. |
Tale |
Hong Kong Hong Kong University of Science and Technology |
Tale is a presentation coach which gives feedback to users rehearsing a presentation or speech. The solution provides a professional artificial intelligence platform to help people rehearse their presentation performance anytime and anywhere. The professional AI considers different presentation dimensions and analyzes the video. After analysis, the AI gives instant feedback to the users, showing analytics results and suggestions to help in preparing for interviews, school presentations, and more. |
innobie |
Hungary Neumann Janos Egyetem |
innobie is a smartphone app to help students understand the curriculum they are reading in textbooks. The app projects a 3D image over 2D illustrations in books, and can help elementary students learn biology, chemistry, geography, history, and more through this interactive platform. |
RealVol |
India IIIT Delhi |
RealVol features a virtual reality based immersive walkthrough for CT and MRI scans. It uses an advanced volumetric algorithm to directly convert 2D CT and MRI scan images to 3D volumes for interactive visualizations by doctors and patients. |
Drugsafe |
India RV College of Engineering |
Drugsafe was developed to provide multi-layered checks to validate genuine drugs and decrease illness from counterfeit drugs. Using an app with three unique checks, drugs are quickly validated as safe or not safe, and the user is immediately notified. |
Practikality |
India Amity International School |
Practikality is a 3-in-1 solution to help individuals with a speech impairment, hearing impairment or visual impairment. VOICE gives an expressive output by closely matching human speech; EASI has a customizable database which allows a person to add personal signs which reduces the time lag between interpretations; VISION has features like face recognition and environmental analysis to make every day navigation easier. |
Beehive Drones |
Indonesia University of Manchester |
Beehive Drones is a project that applies the Internet-of-Things based swarm drone technology as a new approach to solve the global increasing demand for food. A cooperative drones system is used to monitor crops and is accessible to farmers via mobile applications. Using the monitoring data, the system provides suggestions to farmers on their crops: whether they need more fertilizer, pesticide, or they are ready to be harvested. |
Mediated Ear |
Japan
University of Tokyo |
Mediated Ear is software for hearing-impaired individuals to focus on a specific speaker among a multitude of conversations. Mediated Ear can relay specific sounds in audio waveforms through deep learning.
|
Emergensor |
Japan University of Tokyo |
Emergensor is a web service which provides security information for people living in conflict-affected areas through mobile application. It analyzes collective peoples’ behavioral data in the case of emergency and has the potential to automatically analyze the behavioral maps through deep learning to share appropriate information to users. This security information helps provide security to those living in conflict-affected regions. |
ezaki-lab |
Japan National Institute of Technology, Toba College |
ezaki-lab created an Artificial Intelligence and Machine Learning system that can automatically feed surface fish, such as red sea bream and pseudocaranx dentex, through detecting the optimal feeding amount for the fish, and determining how to automatize food dispensing. |
En#22.45km |
Korea
Sejong University |
This project supports emergency room operations by providing a solution to effectively monitor and categorize emergency calls. The system transcribes the caller’s statements and analyzes them in real-time to classify the specific category of the problem. Once classified, the program suggests the appropriate manual for emergency room staff. |
CLASSUM |
Korea Advanced Institute of Science and Technology |
CLASSUM is a UI/UX-based interactive learning platform which provides students with a forum to ask questions. The goal is to increase class participation by utilizing a system through which students are comfortable asking question of teachers and each other. |
PINE. |
Malaysia Universiti Tun Hussein Onn Malaysia |
PINE. is a Machine Learning based non-invasive sensor to measure and categorize pineapples based on their sweetness. The goal is to enable Malaysian farmers to reduce waste and increase efficiency when evaluating their produce by predicting a pineapple’s ripeness before harvesting. |
ERANOI |
Mexico Instituto Tecnologico de Veracruz, Instituto Tecnologico de Culiacan, Universidad Insurgentes |
ERANOI created a monitoring system through a pocket-sized IoT device that measures in real-time the vital signs of patients such as: temperature, humidity, heart rate, blood oxygen and heart electrical activity. RedBeat collects the data and automatically sends it to a dashboard shared with the patient’s doctor, all powered by Azure Cloud. More than just a monitoring system, Redbeat is a smart control system that can help prevent diseases such as heart attacks and strokes. |
DeafKIT |
Moldova Lomonosov Moscow State University |
DeafKIT is an automatic solution for sign-language translation based in neural networks. Translations are provided with video capturing to make communication easier and more effective. |
SochWare |
Nepal
Nepal Engineering College, College of Information Technology and Engineering, Madan Bhandari Memorial College |
SochWare designed a solution called E-Agrovet to help farmers identify plant diseases, suggest mitigation strategies, connect with experts, and get updated with recent agriculture findings. Additionally, the solution will also enable farmers to alert nearby harvesters about their crops with the goal of reducing the need for pesticides in farming. |
Team Sentinel |
New Zealand University of Canterbury, University of Auckland |
Sentinel is an intelligent tool that reimagines the way we manage tank water supplies. The team is working to create a platform that facilitates a toolkit with everything from autonomous supply management, to sustainable tips and forecast insights. With key relationships to industry leaders in property management and agricultural, the team strives to bring an industry-standard IoT platform to meet the real problems faced by companies. |
Hypebeat |
New Zealand University of Auckland |
"Making it" as a musician in the new age of digital music, streaming and social media is just as much about doing what works for your brand and business as it is about great music. Hypebeat helps to show musicians what works so they can achieve their creative vision. |
UniRide |
New Zealand University of Auckland |
UniRide is a phone app that will be available on iOS and Android. It is targeted towards students and helps those frustrated with public transport make the transition to carpooling as a faster, cheaper and more social alternative. UniRide seamlessly facilitates the process of finding, creating and managing a carpool between students. |
Fe Amaan |
Pakistan National University of Sciences and Technology |
Fe Amaan is a wearable belt which regularly monitors fetus health through the use of IoT sensor device which is placed on a mother’s abdomen. The device captures the fetal heart rate and movement and sends it via Bluetooth to a mobile app. The main features of this system are to provide automated analysis of fetal health on a regular basis without harming the mother or child. As the system makes remote monitoring of the fetus possible, it aims to reduce the high rate of intrauterine deaths and stillbirths in Pakistan. |
Wavy |
Poland Politechnika Lodzka |
Wavy is an underwater localizer for scuba divers. The team is creating a Big Data system to analyze and use the information to create diving logs, improve tourism, and support research about the influence of water pressure on human bodies. The system will also send tracking data to diving supervisors. |
VisionX |
Romania Politechnica University Timisoara |
VisionX is a system designed to automatically detect anomalies and diseases encountered anywhere in the human body with radiologist-level accuracy, by analyzing common medical X-ray images with the help of the latest Azure technologies, such as Machine Learning and Azure Functions. The product will provide a much-needed solution for people in areas of the world that lack access to radiology diagnostics while also acting as an assistant tool for the medical experts examining radiographies. |
Coffee Break |
Russia Lomonosov Moscow State University |
The team developed a project for the recognition of varieties and taste characteristics of wine and coffee based on their spectral characteristics using machine learning algorithms. Coffee Break is shooting the spectra of known brands in the laboratory, translating them into vector representation, and training a gradient boosting model. Using this approach, the team hopes to match each brand of goods with its unique digital representation in the form of its spectrum. |
Team 7x |
Singapore
Nanyang Polytechnic School of Engineering |
Team 7x created the ProcubeX, an autonomous, fun and smart learning tool designed to help individuals learn and cope better with dyslexia. Equipped with different interactive features, along with machine learning based lessons, each individual student will be assigned lessons that will suit their strengths, weaknesses and experience. This will also help caretakers and teachers assess dyslexic students’ progress. |
Biolegend | Taiwan | "Bioknee" is a rehabilitation system providing clinical therapists and patients a communications platform during a procedure. It can predict the time to heal through Machine Learning, encouraging the patient to continue rehabilitation. |
BeeConnex |
Thailand King Mongkut’s University of Technology Thonburi |
The BeeConnex team designed and developed an IoT-based solution for beekeeping called the Smart Hive. Their solution captures sounds, images, weight, and humidity from inside the hives, and transmits the data to the cloud. Signals are then processed and classified using techniques including: signal processing, feature extraction and deep learning. Beekeepers can remotely monitor several hives in real-time and can be notified instantly when an abnormal situation arises. |
Proland |
Turkey Koc University |
Proland is a machine learning solution aimed to solve problems of inefficiency for agriculture and agriculture-related activities. The model is built on Azure Machine Learning Studio using historical temperature, precipitation and ton/acre yield values for land, allowing farmers to make the best agriculture choices for their land condition. |
InterviewBot | UK | InterviewBot is a web-based application tailored to aid students with video or physical interviews when applying for jobs. It gives real-time feedback on interview-style questions. Companies can also use this tool to assess their candidates’ performance, with the written transcript allowing employers to dissect the interview in detail. |
BlackLight |
UK University of Abertay |
BlackLight’s project, Firepoint, is a First-Person MR simulator, built around helping firefighters explain and display what their daily work environment is like without putting anyone at risk. The simulator allows users to see through the eyes of a firefighter as they make their way through a multi-story training ground with a variety of different encounters to tackle. The goal is to use this tool to help firefighters in community outreach, recruitment, and training. |
Pengram |
USA University of California, Berkeley |
Pengram is an AR/VR platform which allows engineers from around the world to be holographically ‘teleported’ into a workspace when needed. For example, if an operator repairing a $100,000 medical device needs help, they will be able to wire into the service, and using a Hololens can watch an expert perform the repairs in 3D. |
Boomerang |
USA University of California, Berkeley |
Boomerang is a suite of applications that builds a network of support around asthmatics. The patient-specific mobile application uses an affordable device to detect an out-of-range or in-range inhaler, notifying the user when appropriate. Boomerang builds a more accurate assessment of high-risk environments for asthma attacks--in such scenarios, Boomerang partners with mobile assistants such as Siri or Cortana to preemptively draw attention to the inhaler or reach out to appropriate members of the user’s network, building a community of patient-centered support. |
Loro |
USA Harvard University |
Loro is a platform for providing a smart companion robot for wheelchair users. Loro integrates several technologies including: a universally mountable 360 camera, a laser pointer, a flashlight, an alarm system, and a user-friendly interface. Loro is designed to provide the users with intelligent navigation, health monitoring, social interaction and smart home connectivity. |
Team Prometheus |
USA Arizona State University |
Prometheus combines surveillance drones and concepts of machine learning to detect wildfires while they are still in their early stages. The concept is to leverage the vast amount of wildfire images and videos available on the internet to train a machine learning algorithm to detect the presence of a fire. A drone then flies over parks and forests collecting the images, and an algorithm will determine the confidence level for if a region is about to develop a fire. |
Vinculum |
USA University of Massachusetts |
Vinculum leverages image processing and computer vision technology to determine if separated family members might be in another location and help reunite refugee families. The application determines a similarity score between a submitted image and the photos in the database. The application utilizes Microsoft’s Cognitive Vision APIs and Azure ML models to determine a similarity score between two photos. The user will then be shown the top matches. |
Zelixa |
USA Harvard University California Polytechnic State University, San Luis Obispo |
Zelixa is a Hololens app and online learning platform built to help dyslexic kids learn to read. Aiming to tackle some of the biggest challenges facing dyslexia therapy today, Zelixa is the first app to uniquely combine mixed reality, gamification, and cutting edge scientific research into a tool that is immediately transferable, improvement-focused, and even fun. |
Save the date for the Imagine Cup World Finals on July 25th at 9:00am PST to see which team will become the 2018 Imagine Cup Champions and win up to $100,000!