Implementation of the CVPR 2022 paper DynamicEarthNet: Daily Multi-Spectral Satellite Dataset for Semantic Change Segmentation.
Earth observation is a fundamental tool for monitoring the evolution of land use in specific areas of interest. Observing and precisely defining change, in this context, requires both time-series data and pixel-wise segmentations. To that end, we propose the DynamicEarthNet dataset that consists of daily, multi-spectral satellite observations of 75 selected areas of interest distributed over the globe with imagery from Planet Labs. These observations are paired with pixel-wise monthly semantic segmentation labels of 7 land use and land cover (LULC) classes. DynamicEarthNet is the first dataset that provides this unique combination of daily measurements and high-quality labels. In our experiments, we compare several established baselines that either utilize the daily observations as additional training data (semi-supervised learning) or multiple observations at once (spatio-temporal learning) as a point of reference for future research. Finally, we propose a new evaluation metric SCS that addresses the specific challenges associated with time-series semantic change segmentation.
The code has been implemented & tested with Python 3.9.6 and Pytorch 1.9.0.
This repository only contains the implementation of temporal architectures. For the semi-supervised training, we use Context Aware Consistency repository. We refer the interested reader there. The models are adapted from Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networks (ICCV 2021).
config/defaults.yaml
.utae_train.py
.config/defaults.yaml
.helper/parser.py
.
To test our architecture use the pretained models, given above, and run inference.py
. Modify the checkpoint directory path!
If you use our implementation, please cite:
@InProceedings{Toker_2022_CVPR,
author = {Toker, Aysim and Kondmann, Lukas and Weber, Mark and Eisenberger, Marvin
and Andres, Camero and Hu, Jingliang and Hoderlein, Ariadna and Senaras, Caglar and Davis, Timothy and Cremers, Daniel and Marchisio, Giovanni and Zhu, Xiaoxiang and Leal-Taixe, Laura},
title = {DynamicEarthNet: Daily Multi-Spectral Satellite Dataset for Semantic Change Segmentation},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2022}
}