Google Research has developed StreetReaderAI, a prototype that uses multimodal AI to make Google Street View accessible to visually impaired users. The system leverages Gemini AI to provide real-time audio descriptions of street scenes, intersections, and landmarks. Users can navigate through virtual streetscapes using voice commands or keyboard shortcuts, receiving dynamic feedback about their surroundings and options for movement, similar to an accessible video game experience. AI
IMPACT Enhances accessibility of visual data, potentially opening up vast datasets like Street View to visually impaired users.
RANK_REASON Research paper detailing a new AI prototype for accessibility. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Google AI / Research →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →