Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding

He, Songtao; Bastani, Favyen; Jagwani, Satvat; Alizadeh, Mohammad; Balakrishnan, Hari; Chawla, Sanjay; Elshrif, Mohamed M.; Madden, Samuel; Sadeghi, Amin

doi:10.48550/arxiv.2007.09547

Cited by 5 publications

(5 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also outperform other popular techniques for road network extraction such as DeepRoadMapper [25], D-LinkNet [43], RoadTracer [4] etc. Furthermore, our result is also better than the recent Sat2Graph [17] technique on SpaceNet dataset.…”

Section: Resultsmentioning

confidence: 68%

PP-LinkNet: Improving Semantic Segmentation of High Resolution Satellite Imagery with Multi-stage Training

Tran¹,

Zonoozi²,

Varadarajan³

et al. 2020

Proceedings of the 2nd Workshop on Structuring and Understanding of Multimedia heritAge Contents

View full text Add to dashboard Cite

Road network and building footprint extraction is essential for many applications such as updating maps, traffic regulations, city planning, ride-hailing, disaster response etc. Mapping road networks is currently both expensive and labor-intensive. Recently, improvements in image segmentation through the application of deep neural networks has shown promising results in extracting road segments from large scale, high resolution satellite imagery. However, significant challenges remain due to lack of enough labeled training data needed to build models for industry grade applications. In this paper, we propose a two-stage transfer learning technique to improve robustness of semantic segmentation for satellite images that leverages noisy pseudo ground truth masks obtained automatically (without human labor) from crowd-sourced OpenStreetMap (OSM) data. We further propose Pyramid Pooling-LinkNet (PP-LinkNet), an improved deep neural network for segmentation that uses focal loss, poly learning rate, and context module. We demonstrate the strengths of our approach through evaluations done on three popular datasets over two tasks, namely, road extraction and building footprint detection. Specifically, we obtain 78.19% meanIoU on SpaceNet building footprint dataset, 67.03% and 77.11% on the road topology metric on SpaceNet and DeepGlobe road extraction dataset, respectively. CCS CONCEPTS • Computing methodologies → Computer vision; Neural networks; • Information systems → Data mining.

show abstract

Section: Resultsmentioning

confidence: 68%

PP-LinkNet: Improving Semantic Segmentation of High Resolution Satellite Imagery with Multi-stage Training

Tran¹,

Zonoozi²,

Varadarajan³

et al. 2020

Proceedings of the 2nd Workshop on Structuring and Understanding of Multimedia heritAge Contents

View full text Add to dashboard Cite

show abstract

“…In recent years, line-shaped object detection has drawn great attention. The current works have multiple different goals, including road-network extraction [7], [8], [11], [16], lane detection [9], [10], [12], [17], [18], automatic annotation [19], [20], etc. Among them, some recent works detected line-shaped objects by iterative graph growing (i.e., grow the graph vertex by vertex) [7], [10], [20]- [22], but they are limited to detecting objects with specific structures and suffer from accumulated errors.…”

Section: A Line-shaped Object Detectionmentioning

confidence: 99%

“…In past works about structural predictions, such as roadnetwork extraction [16] and road-lane detection [10], the obtained results are evaluated by pixel-level metrics including precision, recall and F1-score, as well as structureaware metrics, such as APLS [33] and Connectivity [10].…”

Section: Evaluation Metricsmentioning

confidence: 99%

CP-loss: Connectivity-preserving Loss for Road Curb Detection in Autonomous Driving with Aerial Images

Xu¹,

Sun²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Road curb detection is important for autonomous driving. It can be used to determine road boundaries to constrain vehicles on roads, so that potential accidents could be avoided. Most of the current methods detect road curbs online using vehicle-mounted sensors, such as cameras or 3-D Lidars. However, these methods usually suffer from severe occlusion issues. Especially in highly-dynamic traffic environments, most of the field of view is occupied by dynamic objects. To alleviate this issue, we detect road curbs offline using high-resolution aerial images in this paper. Moreover, the detected road curbs can be used to create high-definition (HD) maps for autonomous vehicles. Specifically, we first predict the pixel-wise segmentation map of road curbs, and then conduct a series of post-processing steps to extract the graph structure of road curbs. To tackle the disconnectivity issue in the segmentation maps, we propose an innovative connectivity-preserving loss (CP-loss) to improve the segmentation performance. The experimental results on a public dataset demonstrate the effectiveness of our proposed loss function. This paper is accompanied with a demonstration video and a supplementary document, which are available at https://sites.google.com/view/cp-loss.

show abstract

“…This paper builds off of and is inspired by two main papers, Sat2Graph (He et al 2020) and RoadRunner (He et al 2018).…”

Section: Introductionmentioning

confidence: 99%

Using Reinforcement Learning to Iteratively Construct Road Networks from Satellite Images and GPS Data

Gallardo

2024

AAAI

View full text Add to dashboard Cite

show abstract

Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding

Cited by 5 publications

References 35 publications

PP-LinkNet: Improving Semantic Segmentation of High Resolution Satellite Imagery with Multi-stage Training

PP-LinkNet: Improving Semantic Segmentation of High Resolution Satellite Imagery with Multi-stage Training

CP-loss: Connectivity-preserving Loss for Road Curb Detection in Autonomous Driving with Aerial Images

Using Reinforcement Learning to Iteratively Construct Road Networks from Satellite Images and GPS Data

Contact Info

Product

Resources

About