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

2025-08-12

AI Summary

Limited by the scarcity of large-scale, high-quality annotated data, curb ramp detection in street-view imagery suffers from suboptimal accuracy. To address this, we propose a two-stage automated framework: first, we innovatively leverage government-released geographic metadata of curb ramps to achieve high-precision automatic mapping from geospatial coordinates to pixel-level locations in street-view panoramic images; second, we construct CurbRamp-1M—the first large-scale, high-fidelity benchmark dataset for curb ramp detection. Building upon this dataset, we design a lightweight, improved ConvNeXt V2-based detector optimized end-to-end. Experiments demonstrate that our dataset achieves 94.0% precision and 92.5% recall, while our model attains an AP of 0.9236—significantly outperforming prior methods. This work establishes a scalable foundation—both data and model—for intelligent perception of accessible urban environments.

Technology Category

• Computer Vision: Large Vision Models
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Humans and AI: AI for Accessibility

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data

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.

Problem

• Detecting curb ramps in images lacks large-scale datasets
• Existing datasets suffer from low quality or small scale
• Proposing RampNet to improve dataset scale and detection performance

Innovation

• Two-stage pipeline for curb ramp detection
• Auto-translate government data to annotations
• Modified ConvNeXt V2 model achieves high AP