Keywords
Visual Impairments
Keywords
Visual Impairments
Partners | Funders
Funded by Microsoft AI for Accessibility
Publications
Theodorou, L., Massiceti, D., Zintgraf, L., Stumpf, S., Morrison, C., Cutrell, E., Harris, M. T., & Hofmann, K. (2021). Disability-first Dataset Creation: Lessons from Constructing a Dataset for Teachable Object Recognition with Blind and Low Vision Data Collectors. Proceedings of the 23rd International ACM SIGACCESS Conference on Computers and Accessibility, 1–12. https://doi.org/10.1145/3441852.3471225
Massiceti, D., Theodorou, L., Zintgraf, L., Harris, M. T., Stumpf, S., Morrison, C., … Hofmann, K. (2021). ORBIT: A real-world few-shot dataset for teachable object recognition collected from people who are blind or low vision. https://doi.org/10.25383/city.14294597.v3
Project Summary
Novel smartphones apps using Artificial Intelligence (AI) are useful in making visual information accessible to people who are blind or low vision. However, research into issues such as identifying which objects belong to the user and what things are particularly important to users who are blind or have low vision is held back by the lack of available data, particularly from people who are blind or low vision. The ORBIT (Object Recognition for Blind Image Training) project, funded by Microsoft AI for Accessibility, aimed to construct a large dataset for training and evaluating AI algorithms for personalised object recognition for blind and low vision users.
ORBIT
Object Recognition for Blind Image Training
