TeachTOK is a collaborative interactive machine teaching application for image classification task. You can try TeachTOK online here!
TeachTOK is designed to let a group of people collaborate to train an image classification model asynchronously. In TeachTOK, a group of users constitutes a collective dataset of images to train an image classifier. Users receive feedback on the classifier's performance, they can inspect errors and communicate with each other.
TeachTOK is a web application developed using Marcelle toolkit and redesigned using the SvelteKit framework. Following Marcelle's architecture, it uses a Node.js server associated with a MongoDB database for data storage and synchronization, and a web client written in TypeScript. The server relies on the Feathers.js framework for authentication and data storage.
Runs the app in the development mode. Open http://localhost:5173 to view it in the browser.
The page will reload if you make edits.
Builds a static copy of your site to the dist/ folder.
Your app is ready to be deployed!
TeachTOK is developed for a research project led by Behnoosh Mohammadzadeh (PhD student at LISN and ISIR), Jules Françoise (CNRS researcher at LISN), Baptiste Caramiaux (CNRS researcher (tenured, HDR) at ISIR) and Michèle Gouiffès (Maître de Conférences (HDR) at LISN).
This research was supported by the ARCOL project (ANR-19-CE33-0001) and the ELEMENT project (ANR-18-CE33-0002) from the French National Research Agency.
- Behnoosh Mohammadzadeh (PhD student at LISN and ISIR)
- Jules Françoise (CNRS researcher at LISN)
- Baptiste Caramiaux (CNRS researcher(tenured, HDR) at ISIR)
- Abel Henry-Lapassat (M1 Intern at LISN)
Please cite the following publication when refering to TeachTOK in academic publications:
Mohammadzadeh, Behnoosh, et al. "Studying Collaborative Interactive Machine Teaching in Image Classification." Proceedings of the 29th International Conference on Intelligent User Interfaces. 2024. DOI: 10.1145/3640543.3645204. PDF