Skip to main content

Hello, I'm Olga. I'm a Front-End Developer and PhD candidate.

I often experiment with museum collections using machine learning algorithms.

My recent projects

reGrant

reGrant

regrant.herokuapp.com

A Web-based recommender system that generates personalised museum tours for the UCL Grant Museum of Zoology and Comparative Anatomy, London, UK.

At the end of the visit, the system compiles a visit summary report with information about user preferences and what the user might like to see next time they come to the Grant Museum.

Stack: Python Flask, Angular 8, MongoDB, mLab, d3.js

NHM object recognition app

An app to detect museum objects in real time.

This prototype, developed in Flutter, is aimed to help museum visitors understand what they are looking at without seeking assistance or object labels in the physical museum.

Teachable machine was used to train the custom model with some objects from the Natural History Museum collection available on Sketchfab.

Stack: Flutter / Dart, Tensorflow Lite

Image processor

editor.p5js.org/olgaloboda/sketches/N0XfdbHYd

An experiment to study colour and sound composition of paintings.

Using image processing, this program generates a linear colour gradient from the pixels of selected painting, and then composes an audio file with the sounds that correspond to the pixel colour values.

Stack: JavaScript, p5.js

PoseNet

PoseNet-based sculpture finder

editor.p5js.org/olgaloboda/sketches/9uVQwJ11s

An experiment with a real-time pose estimation to find sculptures similar to your pose.

Please note that you have to enable a Webcam and WebGL in your browser to test this project.

Stack: PoseNet / Tensorflow.js, JavaScript, ml5.js, p5.js

Clone of Apple's Weather app

Similar to the original app, the clone allows to check the weather in selected locations, to add/delete locations from the list and to convert temperatures in Celsius to Fahrenheit.

The project used the following APIs: MetaWeather API for the weather forecast and icons, and Google Places API for the instant search by location feature.

Stack: Flutter / Dart

My research

1. Loboda, O., Nyhan, J., Mahony, S., Romano, D. M. and Terras, M. (2019). 'Content-based Recommender Systems for Heritage: developing a personalised museum tour', in Proceedings of 1st International 'Alan Turing' Conference on Decision Support and Recommender Systems (DSRS-Turing 2019). London, UK. Available at: https://www.researchgate.net/publication/337428660_Content-based_Recommender_Systems_for_Heritage_Developing_a_Personalised_Museum_Tour

2. Loboda, O., Nyhan, J., Mahony, S. and Romano, D. M. (2018). 'Towards Evaluating the Impact of Recommender Systems on Visitor Experience in Physical Museums', in Proceedings of the Mobile Human-Computer Interaction Conference (Mobile Cultural Heritage Workshop). Barcelona, Spain. Available at: https://www.researchgate.net/publication/327319485_Towards_Evaluating_the_Impact_of_Recommender_Systems_on_Visitor_Experience_in_Physical_Museums

3. Kontiza K., Loboda, O., Deladienne, L., Castagnos, S. and Naudet, Y. (2018). 'A Museum App to Trigger Users' Reflection', in Proceedings of the Mobile Human-Computer Interaction Conference (Mobile Cultural Heritage Workshop). Barcelona, Spain.

Peer reviewed papers for:

  1. International Journal of Human-Computer Studies
  2. 8th International Conference on Affective Computing & Intelligent Interaction (ACII) 2019
  3. ACM CHI Conference on Human Factors in Computing Systems 2020
  4. 22nd ACM International Conference on Multimodal Interaction