Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/wgcban/spin_roadmapper
Official implementation of our ICRA'22 paper: SPIN Road Mapper: Extracting Roads from Aerial Images via Spatial and Interaction Space Graph Reasoning for Autonomous Driving
https://github.com/wgcban/spin_roadmapper
aerial-imagery autonomous-driving computer-vision road-detection satellite-imagery segmentation
Last synced: 12 days ago
JSON representation
Official implementation of our ICRA'22 paper: SPIN Road Mapper: Extracting Roads from Aerial Images via Spatial and Interaction Space Graph Reasoning for Autonomous Driving
- Host: GitHub
- URL: https://github.com/wgcban/spin_roadmapper
- Owner: wgcban
- License: apache-2.0
- Created: 2021-09-14T22:09:19.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2024-01-18T13:40:22.000Z (10 months ago)
- Last Synced: 2024-01-29T17:12:33.065Z (10 months ago)
- Topics: aerial-imagery, autonomous-driving, computer-vision, road-detection, satellite-imagery, segmentation
- Language: Jupyter Notebook
- Homepage: https://www.wgcban.com/research/spin-road-mapper
- Size: 16.7 MB
- Stars: 67
- Watchers: 3
- Forks: 14
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
[](https://paperswithcode.com/sota/road-segementation-on-deepglobe?p=spin-road-mapper-extracting-roads-from-aerial)
[](https://paperswithcode.com/sota/road-segementation-on-massachusetts-roads?p=spin-road-mapper-extracting-roads-from-aerial)
## SPIN Road Mapper: Extracting Roads from Aerial Images via Spatial and Interaction Space Graph Reasoning for Autonomous Driving (ICRA'22)
[Wele Gedara Chaminda Bandara](https://www.wgcban.com/), [Jeya Maria Jose Valanarasu](https://jeya-maria-jose.github.io/research/), and [Vishal M. Patel](https://engineering.jhu.edu/vpatel36/sciencex_teams/vishalpatel/)
Read Paper: [Link](https://arxiv.org/abs/2109.07701)
**Accepted** for presentation at the 2022 IEEE International Conference on Robotics and Automation (ICRA), May 23-27, 2022, Philadelphia (PA), USA.
## Overview of proposed SPIN module
We build graphs in two spaces: (a) spatial space and (b) a projected latent interaction space from feature maps. Graph reasoning in spatial space extracts connectivity between the road segments, whereas reasoning over interaction space delineates roads from other topographies. Nodes connected with lines in (a) denote how road segments are modeled to understand connectivity in the spatial space. Regions marked with different colors in (b) denote how different semantics are segregated for better road delineation in the interaction space.
## Architecture of proposed SPIN module and SPIN pyramid
The architecture of our proposed method. (a) We perform graph reasoning in both spatial and interaction space. (b) The proposed SPIN pyramid module which performs SPIN graph reasoning at multiple scales 1, 1/2, and 1/4 of original feature map to extract multi-scale long-range contextual information.
## Proposed network for road segmentation from aerial images
The input images are first feed forwarded to a feature extractor block followed by a bottleneck consisting of stack of two hourglass modules. Then, the output of bottleneck is passed through a segmentation branch which consists of conv layers, our SPIN pyramid and a final classification layer to get the road segmentation map.
## A qualitative comparison between our SPIN Road Mapper and the SOTA methods
## Reproducing the results
### 1. Donwloading the datasets
In this paper we used two publically available road segmentation datasets, namely (1) Massachusetts road dataset, and (2) DeepGlobe dataset.
The Massachusetts road dataset can be downloaded from: [Click Here](https://www.cs.toronto.edu/~vmnih/data/)
## Complete segmentation network with SPIN module
The main module can be found at
modelsstack_module.py/StackHourglassNetMTL_DGCNv4