CoDEx: Combining Domain Expertise for Spatial Generalization in Satellite Image Analysis
This addresses the problem of spatial generalization for satellite image analysis, which is crucial for applications like environmental monitoring, but the approach is incremental as it builds on existing domain generalization techniques.
The paper tackles the challenge of poor model performance in satellite image analysis due to terrain variations across locations by proposing a domain-generalization framework that trains expert models per training domain and aggregates their predictions, achieving consistent gains over existing methods on four datasets.
Global variations in terrain appearance raise a major challenge for satellite image analysis, leading to poor model performance when training on locations that differ from those encountered at test time. This remains true even with recent large global datasets. To address this challenge, we propose a novel domain-generalization framework for satellite images. Instead of trying to learn a single generalizable model, we train one expert model per training domain, while learning experts' similarity and encouraging similar experts to be consistent. A model selection module then identifies the most suitable experts for a given test sample and aggregates their predictions. Experiments on four datasets (DynamicEarthNet, MUDS, OSCD, and FMoW) demonstrate consistent gains over existing domain generalization and adaptation methods. Our code is publicly available at https://github.com/Abhishek19009/CoDEx.