Our findings indicate that adopting "advanced" computational elements fails to significantly improve registration accuracy. Instead, well-established registration-specific designs offer fair improvements, enhancing results by a marginal 1.5\% over the baseline. Our findings emphasize the importance of rigorous, unbiased evaluation and contribution disentanglement of all low- and high-level registration components, rather than simply following the computer vision trends with "more advanced" computational blocks. We advocate for simpler yet effective solutions and novel evaluation metrics that go beyond conventional registration accuracy, warranting further research across diverse organs and modalities. The code is available at \url{https://github.com/BailiangJ/rethink-reg}.
@article{arxiv.2407.19274,
title = {Mamba? Catch The Hype Or Rethink What Really Helps for Image Registration},
author = {Bailiang Jian and Jiazhen Pan and Morteza Ghahremani and Daniel Rueckert and Christian Wachinger and Benedikt Wiestler},
journal= {arXiv preprint arXiv:2407.19274},
year = {2025}
}
Comments
WBIR 2024 Workshop on Biomedical Imaging Registration