RampNet: A Two-Stage Pipeline for Bootstrapping Curb Ramp Detection in Streetscape Images from Open Government Metadata

TL;DR

This paper presents RampNet, a two-stage pipeline that automatically creates a large-scale, high-quality curb ramp dataset from government data and trains a model achieving state-of-the-art detection performance in streetscape images.

Contribution

It introduces the first large-scale, high-quality curb ramp dataset and a novel two-stage pipeline for dataset generation and model training.

Findings

Generated over 210,000 annotated panoramas with 94.0% precision and 92.5% recall.

Achieved state-of-the-art detection with 0.9236 AP.

Outperformed prior methods significantly.

Abstract

Curb ramps are critical for urban accessibility, but robustly detecting them in images remains an open problem due to the lack of large-scale, high-quality datasets. While prior work has attempted to improve data availability with crowdsourced or manually labeled data, these efforts often fall short in either quality or scale. In this paper, we introduce and evaluate a two-stage pipeline called RampNet to scale curb ramp detection datasets and improve model performance. In Stage 1, we generate a dataset of more than 210,000 annotated Google Street View (GSV) panoramas by auto-translating government-provided curb ramp location data to pixel coordinates in panoramic images. In Stage 2, we train a curb ramp detection model (modified ConvNeXt V2) from the generated dataset, achieving state-of-the-art performance. To evaluate both stages of our pipeline, we compare to manually labeled panoramas. Our generated dataset achieves 94.0% precision and 92.5% recall, and our detection model reaches 0.9236 AP -- far exceeding prior work. Our work contributes the first large-scale, high-quality curb ramp detection dataset, benchmark, and model.