StreetNav: Leveraging Street Cameras to Support Precise Outdoor Navigation for Blind Pedestrians

TL;DR

This work tackles the challenge of outdoor navigation for BLV individuals, where GPS lacks precision. It proposes StreetNav, a system that repurposes existing street cameras, using a real-time computer vision pipeline and a companion smartphone app to provide precise routing, obstacle awareness, and crossing guidance without storing video data. A community-driven privacy study informs design choices, and both a user study and a technical evaluation demonstrate substantial gains over GPS-based baselines in routing precision, veering reduction, and crossing confidence, though performance degrades with distance and occlusions. The findings highlight the potential to embed accessibility into urban infrastructure while underscoring privacy, scalability, and interoperability challenges that must be addressed for widespread deployment. Overall, StreetNav offers a promising path toward more capable, environment-aware outdoor navigation for BLV people and suggests directions for integrating street-camera insights with existing GPS-based systems.

Abstract

Blind and low-vision (BLV) people rely on GPS-based systems for outdoor navigation. GPS's inaccuracy, however, causes them to veer off track, run into obstacles, and struggle to reach precise destinations. While prior work has made precise navigation possible indoors via hardware installations, enabling this outdoors remains a challenge. Interestingly, many outdoor environments are already instrumented with hardware such as street cameras. In this work, we explore the idea of repurposing existing street cameras for outdoor navigation. Our community-driven approach considers both technical and sociotechnical concerns through engagements with various stakeholders: BLV users, residents, business owners, and Community Board leadership. The resulting system, StreetNav, processes a camera's video feed using computer vision and gives BLV pedestrians real-time navigation assistance. Our evaluations show that StreetNav guides users more precisely than GPS, but its technical performance is sensitive to environmental occlusions and distance from the camera. We discuss future implications for deploying such systems at scale.

Figures (17)

• Figure 1: Street camera used for StreetNav's development and evaluation. The camera is (a) mounted on the building's second floor and (b) faces a four-way intersection.
• Figure 2: Gesture-based localization for determining a user's position on the map. (a) A study participant (P1) is (c) prompted to wave one hand above their head, enabling the computer vision pipeline to distinguish them from other pedestrians in (b) the camera feed view and (d) the map.
• Figure 3: StreetNav's internal graph representation for route planning. The user's current position is added dynamically as a start node to the graph upon choosing a destination. The shortest path, highlighted in green, is then calculated as per this graph representation.
• Figure 4: Identifying obstacles in the user's vicinity. (a) A vehicle turning left yields to the BLV pedestrian (detected in purple) crossing the street. (b) StreetNav identifies the obstacles' category and relative location on the map to provide real-time feedback via the app.
• Figure 5: Recognizing pedestrian signal states. StreetNav compares the number of white and red pixels in the signal crops to determine its state: (a) walk vs. (b) wait.