# RoadGIE: Towards A Global-Scale Aerial Benchmark for Generalizable Interactive Road Extraction

arXiv: https://arxiv.org/abs/2605.26862 · HTML: https://mlsanity.com/paper/2605.26862/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 51 references that are in the map.
- First posted on arXiv 26 May 2026.

## Builds on (most cited first, 30 of 35)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Segment Anything in High Quality](https://mlsanity.com/paper/2306.01567/) (2023)
- [clDice -- A Novel Topology-Preserving Loss Function for Tubular Structure Segmentation](https://mlsanity.com/paper/2003.07311/) (2020)
- [Segment Anything Model for Medical Image Analysis: an Experimental Study](https://mlsanity.com/paper/2304.10517/) (2023)
- [Fast Segment Anything](https://mlsanity.com/paper/2306.12156/) (2023)
- [DeepGlobe 2018: A Challenge to Parse the Earth through Satellite Images](https://mlsanity.com/paper/1805.06561/) (2018)
- [LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation](https://mlsanity.com/paper/2110.08733/) (2021)
- [EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything](https://mlsanity.com/paper/2312.00863/) (2023)
- [SpaceNet: A Remote Sensing Dataset and Challenge Series](https://mlsanity.com/paper/1807.01232/) (2018)
- [Strip Pooling: Rethinking Spatial Pooling for Scene Parsing](https://mlsanity.com/paper/2003.13328/) (2020)
- [RoadTracer: Automatic Extraction of Road Networks from Aerial Images](https://mlsanity.com/paper/1802.03680/) (2018)
- [Dynamic Snake Convolution based on Topological Geometric Constraints for Tubular Structure Segmentation](https://mlsanity.com/paper/2307.08388/) (2023)
- [Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding](https://mlsanity.com/paper/2007.09547/) (2020)
- [Scribble2Label: Scribble-Supervised Cell Segmentation via Self-Generating Pseudo-Labels with Consistency](https://mlsanity.com/paper/2006.12890/) (2020)
- [ScribblePrompt: Fast and Flexible Interactive Segmentation for Any Biomedical Image](https://mlsanity.com/paper/2312.07381/) (2023)
- [Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures](https://mlsanity.com/paper/2404.03010/) (2024)
- [nnInteractive: Redefining 3D Promptable Segmentation](https://mlsanity.com/paper/2503.08373/) (2025)
- [Segment Anything Model for Medical Images?](https://mlsanity.com/paper/2304.14660/) (2023)
- [RobustSAM: Segment Anything Robustly on Degraded Images](https://mlsanity.com/paper/2406.09627/) (2024)
- [Directional Connectivity-based Segmentation of Medical Images](https://mlsanity.com/paper/2304.00145/) (2023)
- [ScribFormer: Transformer Makes CNN Work Better for Scribble-based Medical Image Segmentation](https://mlsanity.com/paper/2402.02029/) (2024)
- [A skeletonization algorithm for gradient-based optimization](https://mlsanity.com/paper/2309.02527/) (2023)
- [ConnNet: A Long-Range Relation-Aware Pixel-Connectivity Network for Salient Segmentation](https://mlsanity.com/paper/1804.07836/) (2018)
- [PRISM: A Promptable and Robust Interactive Segmentation Model with Visual Prompts](https://mlsanity.com/paper/2404.15028/) (2024)
- [Strip R-CNN: Large Strip Convolution for Remote Sensing Object Detection](https://mlsanity.com/paper/2501.03775/) (2025)
- [ScribbleSeg: Scribble-based Interactive Image Segmentation](https://mlsanity.com/paper/2303.11320/) (2023)
- [Towards Satellite Image Road Graph Extraction: A Global-Scale Dataset and A Novel Method](https://mlsanity.com/paper/2411.16733/) (2024)
- [Ret3D: Rethinking Object Relations for Efficient 3D Object Detection in Driving Scenes](https://mlsanity.com/paper/2208.08621/) (2022)
- [From Pixels to Progress: Generating Road Network from Satellite Imagery for Socioeconomic Insights in Impoverished Areas](https://mlsanity.com/paper/2406.11282/) (2024)

## Abstract (arXiv)

Accurate road segmentation from aerial imagery is fundamental to many geospatial applications. However, existing datasets often suffer from limited scene diversity, low semantic granularity, and poor structural continuity, restricting their generalization across environments. To address these challenges, we introduce WorldRoadSeg-360K, the largest and most diverse road segmentation dataset to date, comprising 366,947 high-resolution images collected from 38 countries and 223 cities across various terrains and continents. WorldRoadSeg-360K serves as a comprehensive benchmark and reveals key challenges in handling diverse and structurally complex scenes. Automated approaches often struggle to preserve road connectivity, while current interactive methods lack efficient, topology-sensitive tools for real-world road editing. To this end, we present RoadGIE, establishing a novel interactive paradigm for road extraction in remote sensing. Unlike prior point- or box-based prompting strategies, RoadGIE supports connectivity-aware prompts, including clicks and scribbles, which inherently align with the topology of road networks. To improve structural consistency and mitigate performance degradation during iterative interactions, RoadGIE integrates an expert-guided prompting strategy and adapts the skeleton-based recall loss for interactive scenarios. RoadGIE achieves state-of-the-art performance in both segmentation accuracy and topological consistency on WorldRoadSeg-360K and other benchmarks, while maintaining efficient operation with only 3.7M parameters. The code are publicly available at: https://github.com/chaineypung/RoadGIE