fig2text

URL: https://labs.tib.eu/fig2text

Fig2text is a prototype assistive technology for blind and visually impaired users, aimed to improve accessibility of Open Educational Resources (OER). Learning and teaching materials often use images (photos, maps, diagrams, figures) and thus, only provide limited accessibility for above user groups. In specialised settings, they would be accompanied by high-quality image descriptions to make them accessible, a feature which is missing in most materials present on the Web.

The prototype combines structured description templates and state-of-the-art machine learning techniques to automatically generate image descriptions for commonly used types of educational images. The current functionality includes support for bar and pie charts and will be extended in the future.

Related projects

This work is financially supported by the German Federal Ministry of Education and Research (BMBF) and the European Social Fund (ESF) (InclusiveOCW project, no. 01PE17004).

https://www.tib.eu/en/research-development/project-overview/project-summary/inclusiveocw