1. This paper proposes a novel general change detection architecture, MetaChanger, which includes a series of alternative interaction layers in the feature extractor.
2. Two derived models, ChangerAD and ChangerEx with simple interaction strategies are proposed to verify the effectiveness of MetaChanger.
3. A flow dual-alignment fusion (FDAF) module is proposed for better alignment of bi-temporal features.
The article is generally reliable and trustworthy as it provides detailed information about the proposed change detection architecture, MetaChanger, and its two derived models, ChangerAD and ChangerEx. The article also provides evidence for the effectiveness of these models by citing results from experiments conducted on different scale change detection datasets. However, there are some potential biases that should be noted. For example, the article does not provide any counterarguments or explore any possible risks associated with using these models for change detection tasks. Additionally, the article does not present both sides equally; instead it focuses solely on the benefits of using these models without exploring any potential drawbacks or limitations. Furthermore, there is no mention of any ethical considerations when using these models for change detection tasks such as privacy concerns or data security issues. Finally, there is no discussion about how this technology could be used in other applications beyond remote sensing images or how it could be improved upon in future research efforts.