A comprehensive and trustworthy benchmark of AI methods for change detection in Earth observation
This work presents a standardized, open-source benchmark for evaluating state-of-the-art (SOTA) deep learning methods for Earth observation change detection, and reveals that well-optimized classical architectures, such as Siamese U-Nets, frequently outperform more complex contemporary models when computational efficiency is factored in.
Tadej Tomanič, Alice Baudhuin, Jan Sotošek et al.
· 0 citations