Promoting Connectivity of Network-Like Structures by Enforcing Region Separation

• URL: http://arxiv.org/abs/2009.07011v1
• Date: Tue, 15 Sep 2020 12:21:35 GMT
• Title: Promoting Connectivity of Network-Like Structures by Enforcing Region Separation
• Authors: Doruk Oner and Mateusz Kozi\'nski and Leonardo Citraro and Nathan C. Dadap and Alexandra G. Konings and Pascal Fua
• Abstract summary: We propose a connectivity-oriented loss function for training deep convolutional networks to reconstruct network-like structures. The main idea behind our loss is to express the connectivity of roads, or canals, in terms of disconnections that they create between background regions of the image. We show, in experiments on two standard road benchmarks and a new data set of irrigation canals, that convnets trained with our loss function recover road connectivity so well.
• Score: 101.10228007363673
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose a novel, connectivity-oriented loss function for training deep convolutional networks to reconstruct network-like structures, like roads and irrigation canals, from aerial images. The main idea behind our loss is to express the connectivity of roads, or canals, in terms of disconnections that they create between background regions of the image. In simple terms, a gap in the predicted road causes two background regions, that lie on the opposite sides of a ground truth road, to touch in prediction. Our loss function is designed to prevent such unwanted connections between background regions, and therefore close the gaps in predicted roads. It also prevents predicting false positive roads and canals by penalizing unwarranted disconnections of background regions. In order to capture even short, dead-ending road segments, we evaluate the loss in small image crops. We show, in experiments on two standard road benchmarks and a new data set of irrigation canals, that convnets trained with our loss function recover road connectivity so well, that it suffices to skeletonize their output to produce state of the art maps. A distinct advantage of our approach is that the loss can be plugged in to any existing training setup without further modifications.

Related papers

• Holistically-Nested Structure-Aware Graph Neural Network for Road Extraction [6.537257686406911]
  This paper presents a novel graph neural network (GNN) which simultaneously detects both road regions and road borders. Experiments on challenging dataset demonstrate that the proposed architecture can improve the road border delineation and road extraction accuracy.
  arXiv  Detail & Related papers  (2024-07-02T20:11:09Z)
• Landscaping Linear Mode Connectivity [76.39694196535996]
  linear mode connectivity (LMC) has garnered interest from both theoretical and practical fronts. We take a step towards understanding it by providing a model of how the loss landscape needs to behave topographically for LMC.
  arXiv  Detail & Related papers  (2024-06-24T03:53:30Z)
• Fine-Grained Extraction of Road Networks via Joint Learning of Connectivity and Segmentation [5.496893845821393]
  Road network extraction from satellite images is widely applicated in intelligent traffic management and autonomous driving fields. The high-resolution remote sensing images contain complex road areas and distracted background, which make it a challenge for road extraction. We present a stacked multitask network for end-to-end segmenting roads while preserving connectivity correctness.
  arXiv  Detail & Related papers  (2023-12-07T22:57:17Z)
• Jointly Contrastive Representation Learning on Road Network and Trajectory [11.613962590641002]
  Road network and trajectory representation learning are essential for traffic systems. Most existing methods only contrast within the same scale, i.e., treating road network and trajectory separately. We propose a unified framework that jointly learns the road network and trajectory representations end-to-end.
  arXiv  Detail & Related papers  (2022-09-14T03:08:20Z)
• Topology Preserving Local Road Network Estimation from Single Onboard Camera Image [128.881857704338]
  This paper aims at extracting the local road network topology, directly in the bird's-eye-view (BEV) The only input consists of a single onboard, forward looking camera image. We represent the road topology using a set of directed lane curves and their interactions, which are captured using their intersection points.
  arXiv  Detail & Related papers  (2021-12-19T14:25:22Z)
• Road Network Guided Fine-Grained Urban Traffic Flow Inference [108.64631590347352]
  Accurate inference of fine-grained traffic flow from coarse-grained one is an emerging yet crucial problem. We propose a novel Road-Aware Traffic Flow Magnifier (RATFM) that exploits the prior knowledge of road networks. Our method can generate high-quality fine-grained traffic flow maps.
  arXiv  Detail & Related papers  (2021-09-29T07:51:49Z)
• Hierarchical Recurrent Attention Networks for Structured Online Maps [91.28820076955128]
  We tackle the problem of online road network extraction from sparse 3D point clouds. Our method is inspired by how annotator builds a lane graph, by first identifying how many lanes there are and then drawing each one in turn. We develop a hierarchical recurrent network that attends to initial regions of a lane boundary and traces them out completely by outputting a structured polyline.
  arXiv  Detail & Related papers  (2020-12-22T19:35:53Z)
• We Learn Better Road Pothole Detection: from Attention Aggregation to Adversarial Domain Adaptation [22.076261078410752]
  Road pothole detection results are always subjective, because they depend entirely on the individual experience. Our recently introduced disparity (or inverse depth) transformation algorithm allows better discrimination between damaged and undamaged road areas. We propose a novel attention aggregation (AA) framework, which takes the advantages of different types of attention modules.
  arXiv  Detail & Related papers  (2020-08-16T05:17:42Z)
• TopoAL: An Adversarial Learning Approach for Topology-Aware Road Segmentation [56.353558147044]
  We introduce an Adversarial Learning (AL) strategy tailored for our purposes. We use a more sophisticated discriminator that returns a label pyramid describing what portions of the road network are correct. We will show that it outperforms state-of-the-art ones on the challenging RoadTracer dataset.
  arXiv  Detail & Related papers  (2020-07-17T16:06:45Z)
• RoadTagger: Robust Road Attribute Inference with Graph Neural Networks [26.914950002847863]
  Road attributes such as lane count and road type are difficult to infer from satellite imagery. RoadTagger is an end-to-end architecture which combines Convolutional Neural Networks (CNNs) and Graph Neural Networks (GNNs) to infer road attributes. We evaluate RoadTagger on both a large real-world dataset covering 688 km2 area in 20 U.S. cities and a synthesized micro-dataset.
  arXiv  Detail & Related papers  (2019-12-28T06:09:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.