Student`s projects:
Development of a Deep Learning-based Body Posture Correction Model
Описание проекта:
Due to the fact that we spend more time sitting (at the office/school desk, public transport, etc.), body posture is affected. This contributes to poor body posture such as slumping, slouching, and misalignment in the spinal cord.
The use persuasive technology as well as gamification for reinforcing positive attitudes in people has been adopted by clinicians, psychologists and healthcare practitioners.
The goal of this project is to develop a tool that helps people train their posture each day for a number of minutes.
The underlying technology is built on computer vision modelling using pose estimator neural network models.
Цель:
The goal is to build an end-to-end deep learning (computer vision) model and integrate into a web application for correcting body posture. Students will obtain new skills in end-to-end deep learning in the area of projects related to computer vision.
Критерии оценки:
Students must:
· Have a basic knowledge in Python for Machine Learning
· Understand basic deep neural networks for computer vision
· Understand basic MLOps for deploying ML models on web applications (Python “Streamlit”)
Результаты:
The end-product of this project will be to realize a deep learning model trained with computer vision techniques and integrated into web platform. The model can also be adopted into a mobile applications.
Applying Multi-Criteria Decision Making to Prioritization of Web 3.0 Development Factors
Project description:
Web technologies built on blockchain-based technology, popularly known as Web 3.0 is currently making headlines. Researchers, investors, and startups are pushing resources into the next age of the web due to the decentralized nature of Web 3.0. Web development (user experience and user interface) are of primary focus and as such when it comes to selecting a blockchain framework to build a web platform, it is important for developers to consider certain factors.
The goal of this study is to understand the factors relevant to building a blockchain-based web platform and rank them in order of priority.
Target:
The concept of blockchain has taken the world by storm, yet blockchain is not only limited to cryptocurrency.
In recent months there have been talks about the metaverse, decentralized finance (DeFi), decentralized web applications (DApps), etc. All these are components of the paradigm known as Web 3.0.
In developing such web-based systems for the future, this project seeks to outline the factors to be considered when developing Web 3.0 solutions.
In order to prioritize these factors, the project aims at applying Multi-Criteria Decision Making (MCDM).
Criteria for evaluation:
Students must:
· Understand multi-criteria decision making and analysis concepts
- Understand methods and approaches of qualitative data collection
- Have a basic understanding of blockchain technology
Be able to conduct scientific research and prepare publications.
Results:
Scientific Publication of a qualitative and quantitative analytical study for the application of MCDM to Web 3.0 development.
Publication URL: Applying Multi-Criteria Decision Making to Prioritization of Web 3.0 Development Factors
