RoadTracer: Automatic Extraction of Road Networks from Aerial Images

Bastani, Favyen; He, Songtao; Abbar, Sofiane; Alizadeh, Mohammad; Balakrishnan, Hari; Chawla, Sanjay; Madden, Sam; DeWitt, David J.

doi:10.1109/cvpr.2018.00496

Cited by 296 publications

(287 citation statements)

References 19 publications

Supporting

Mentioning

287

Contrasting

Order By: Relevance

“…• GraphRNN-2D [41,7]: We enhance the GraphRNN model by introducing extra branches to encode/decode node coordinates and city id. We add a CNN that takes into account local rendering of existing graph as in [7] and add checks to avoid invalid edge crossing during inference.…”

Section: Resultsmentioning

confidence: 99%

“…Urban GraphRNN-2D [41,7] PGGAN [23] CityEngine PGGAN is unable to create new cities by either severely overfitting, or producing artifacts. CityEngine produces less style richness due to its fixed rule-based synthesis algorithm.…”

Section: Metricsmentioning

confidence: 99%

“…SpaceNet Dataset. While several datasets have been presented for road detection [2,14,38,7], we use SpaceNet [2] for its large scale, image quality, and open license. To facilitate consistent future benchmarking, we reformat the raw data into an easy-to-use version with consistent tile size in metric space.…”

Section: Satellite Road Parsingmentioning

confidence: 99%

See 2 more Smart Citations

Neural Turtle Graphics for Modeling City Road Layouts

Chu

Li²,

Acuna

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

Figure 1: We introduce Neural Turtle Graphics (NTG), a deep generative model for planar graphs. In the Figure, we show NTG's applications to (interactive) city road layout generation and parsing. AbstractWe propose Neural Turtle Graphics (NTG), a novel generative model for spatial graphs, and demonstrate its applications in modeling city road layouts. Specifically, we represent the road layout using a graph where nodes in the graph represent control points and edges in the graph represents road segments. NTG is a sequential generative model parameterized by a neural network. It iteratively generates a new node and an edge connecting to an existing node conditioned on the current graph. We train NTG on Open Street Map data and show that it outperforms existing approaches using a set of diverse performance metrics. Moreover, our method allows users to control styles of generated road layouts mimicking existing cities as well as to sketch parts of the city road layout to be synthesized. In addition to synthesis, the proposed NTG finds uses in an analytical task of aerial road parsing. Experimental results show that it achieves state-of-the-art performance on the SpaceNet dataset.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Neural Turtle Graphics for Modeling City Road Layouts

Chu

Li²,

Acuna

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

show abstract

“…Rather than segmenting the imagery, RoadTracer [2] and IDL [17] employ an iterative graph construction (IGC) approach that extracts roads via a series of steps in a search process. On each step, a CNN is queried to determine what direction to move in the search, and a road segment is added to a partial road network graph in that direction.…”

Section: Related Workmentioning

confidence: 99%

“…Fundamentally, high error rates make full automation impractical. Even state-of-the-art automatic map inference approaches have error rates between 5% and 10% [2,15]. Navigating the road network using road maps with such high frequencies of errors would be virtually impossible.…”

Section: Introductionmentioning

confidence: 99%

Machine-assisted map editing

Bastani¹,

He²,

Abbar³

et al. 2018

Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems

Self Cite

View full text Add to dashboard Cite

Mapping road networks today is labor-intensive. As a result, road maps have poor coverage outside urban centers in many countries. Systems to automatically infer road network graphs from aerial imagery and GPS trajectories have been proposed to improve coverage of road maps. However, because of high error rates, these systems have not been adopted by mapping communities. We propose machine-assisted map editing, where automatic map inference is integrated into existing, human-centric map editing workflows. To realize this, we build Machine-Assisted iD (MAiD), where we extend the web-based OpenStreetMap editor, iD, with machine-assistance functionality. We complement MAiD with a novel approach for inferring road topology from aerial imagery that combines the speed of prior segmentation approaches with the accuracy of prior iterative graph construction methods. We design MAiD to tackle the addition of major, arterial roads in regions where existing maps have poor coverage, and the incremental improvement of coverage in regions where major roads are already mapped. We conduct two user studies and find that, when participants are given a fixed time to map roads, they are able to add as much as 3.5x more roads with MAiD.

show abstract

Extracting roads from satellite images via enhancing road feature investigation in learning

Feng,

Hou,

Chen

et al. 2024

Computer Animation & Virtual

View full text Add to dashboard Cite

It is a hot topic to extract road maps from satellite images. However, it is still very challenging with existing methods to achieve high‐quality results, because the regions covered by satellite images are very large and the roads are slender, complex and only take up a small part of a satellite image, making it difficult to distinguish roads from the background in satellite images. In this article, we address this challenge by presenting two modules to more effectively learn road features, and so improving road extraction. The first module exploits the differences between the patches containing roads and the patches containing no road to exclude the background regions as many as possible, by which the small part containing roads can be more specifically investigated for improvement. The second module enhances feature alignment in decoding feature maps by using strip convolution in combination with the attention mechanism. These two modules can be easily integrated into the networks of existing learning methods for improvement. Experimental results show that our modules can help existing methods to achieve high‐quality results, superior to the state‐of‐the‐art methods.

show abstract

RoadTracer: Automatic Extraction of Road Networks from Aerial Images

Cited by 296 publications

References 19 publications

Neural Turtle Graphics for Modeling City Road Layouts

Neural Turtle Graphics for Modeling City Road Layouts

Machine-assisted map editing

Extracting roads from satellite images via enhancing road feature investigation in learning

Contact Info

Product

Resources

About