Toolkit for xView2 Challenge: Assessing Building Damage caused by Natural Disasters
-
Updated
Jun 28, 2022 - Jupyter Notebook
Toolkit for xView2 Challenge: Assessing Building Damage caused by Natural Disasters
SpaceNet5 and xView2 code.
A PyTorch Deep Learning pipeline utilizing a U-Net (ResNet34 backbone) to perform semantic segmentation of post-disaster satellite imagery for building damage mapping.
Siamese U-Net with ResNet34 encoder for pre/post-event damage classification and segmentation on xView2.
Semantic segmentation with convolutional neural networks (CNNs) and zoomout features in PyTorch.
Auto Assess Building Damage Based on Siamese Neural Network
🖥 A deployed streamlit application for Alivio
[NeurIPS-REO 2026 Oral] Shortcut learning in geospatial foundation models for cross-disaster damage assessment.
Streamlit disaster damage detection prototype using Roboflow and satellite imagery.
Change detection in pre- and post-disaster satellite imagery using custom CNNs and pretrained models for damage assessment and cross-disaster transfer analysis.
DisasterVision: SSD-Based Detection of Building Damage in Disasters
To associate your repository with the xview2 topic, visit your repo's landing page and select "manage topics."