A simple, strong baseline for building damage detection on the xBD dataset
This work provides a simpler baseline for building damage detection, but it is incremental as it builds on existing methods and highlights generalization challenges in disaster response datasets.
The paper tackled building damage detection by simplifying a complex competition-winning method to create a strong baseline, finding that both the original and simplified models fail to generalize to unseen locations due to dataset issues like unequal class distributions.
We construct a strong baseline method for building damage detection by starting with the highly-engineered winning solution of the xView2 competition, and gradually stripping away components. This way, we obtain a much simpler method, while retaining adequate performance. We expect the simplified solution to be more widely and easily applicable. This expectation is based on the reduced complexity, as well as the fact that we choose hyperparameters based on simple heuristics, that transfer to other datasets. We then re-arrange the xView2 dataset splits such that the test locations are not seen during training, contrary to the competition setup. In this setting, we find that both the complex and the simplified model fail to generalize to unseen locations. Analyzing the dataset indicates that this failure to generalize is not only a model-based problem, but that the difficulty might also be influenced by the unequal class distributions between events. Code, including the baseline model, is available under https://github.com/PaulBorneP/Xview2_Strong_Baseline