The first Workshop on Vision Foundation Models and Generative AI for Accessibility (CV4A11y) took place at ICCV 2025 in Honolulu on October 20, 2025. Co-organized by Jon Froehlich, OSCUR’s PI at the University of Washington, and UW PhD student Chu Li, it brought together the computer vision, HCI and accessibility communities, with 16 accepted papers and six invited speakers.

Two OSCUR papers were presented there. RampNet, led by John O’Meara, then a high school student, turns open government curb ramp inventories into labels on more than 210,000 Street View panoramas and trains a curb ramp detector on them. “Does the cafe entrance look accessible? Where is the door?” outlines geospatial AI agents that answer visual questions about places.