AI Tool NaviSense: Real-Time Object Detection for Visually Impaired Users

Описание: Discover how NaviSense, a new AI-powered smartphone app from Penn State, uses large language models (LLMs) and vision-language models (VLMs) to help visually impaired users quickly find everyday objects in real time.
In this video, you’ll learn:
How NaviSense turns spoken prompts into real-time object detection and guidance.
How audio and vibration cues help users locate objects faster and more accurately than existing tools.
Why using cloud-based LLMs and VLMs removes the need for preloaded object models and increases flexibility.
How interviews with visually impaired users shaped key features like conversational follow‑up questions and hand‑guidance.
What early tests with 12 participants revealed about speed, accuracy, and user experience compared with commercial options.
What’s next for NaviSense, including battery optimization and preparation for commercial release.
This video is ideal for accessibility advocates, AI researchers, app developers, and anyone interested in assistive technology for blind and visually impaired communities.
Keywords: AI for accessibility, object detection app, assistive technology, NaviSense, visually impaired users, real-time object recognition, LLMs, VLMs, Penn State research, smartphone accessibility.
Watch to see how cutting-edge AI can make everyday navigation more independent and empowering for people with visual impairments.