Researchers at Penn State have developed NaviSense, an innovative smartphone application designed to assist visually impaired individuals in navigating their surroundings. Utilizing artificial intelligence (AI) technologies, including large-language and vision-language models, NaviSense recognizes objects in real-time based on voice commands, eliminating the need for preloaded data and enhancing user flexibility. Test users reported a significantly improved experience compared to existing visual aids. The tool’s development was informed by feedback from the visually impaired community, ensuring it meets their specific needs. NaviSense effectively filters out irrelevant objects, asking follow-up questions when needed to refine searches. This groundbreaking tool was recently showcased at the ACM SIGACCESS ASSETS ’25 conference, where it received the Best Audience Choice Poster Award. By revolutionizing the way visually impaired users interact with their environment, NaviSense represents a significant advancement in AI-powered navigation solutions, addressing privacy concerns associated with in-person support options and improving overall accessibility.
Source link
Share
Read more