XClose

In2Research Journeys

Menu

Week 1- Introductory Week + UCL ARC Digital Research Festival a

19 Jun 2026 - Mohamed Mohamoud - Learning, Python, introduction, festival, Research, git

This post marks the completion of my first week as an In2Research intern and the start of many more learning experiences over the next 8 weeks.

Project Overview

My project title is “Data Preprocessing and Benchmarking of Individual AI Models for Eye Image Segmentation”. This project focuses on preprocessing and preparing provided ocular image datasets, including MOBIUS, OpenEDS and Synthetic RIT-Eyes, for eye image segmentation. The aim is to train and benchmark segmentation models, mainly U-Net, with potential real-world applications in areas such as medical image analysis, biometric recognition and gaze tracking.

Highlights of This Week

  • On Tuesday, I attended the UCL ARC Festival. It was a really insightful and action-packed day that gave me a better understanding of the exciting research taking place across ARC. I also had the opportunity to meet and speak with people from a range of backgrounds and career stages, including professors, postdocs and PhD students. I particularly enjoyed talking to PhD students about their research, viewing the poster sessions, and learning more about the different projects being carried out. I also attended several talks presented by ARC members and partners.
  • Set up my developer environment and VS Code, practised Git basics such as git config, cloning, committing, pulling and pushing changes, and started running the repository commands from the documentation, including the initial setup steps and data preparation sections.
  • Started reading papers on ocular datasets, federated learning for diagnosing AMD and segmentation, while also getting more familiar with the repository structure and working through some early setup errors.

Goals for the Placement

  • Build a clear and reproducible preprocessing pipeline.
  • Train and benchmark U-Net on prepared ocular image datasets.
  • Improve my Python, Git and machine learning skills.
  • Become more confident reading and discussing AI/computer vision research papers.
  • Develop better research communication and teamwork skills through meetings and documentation.
  • If time allows, explore additional models or evaluation methods beyond the main U-Net pipeline.