Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Lin, He, Yue, Zhang, Fengting, Wang, Yaonan, Lin, Fengming, Chen, Xiang, Liu, Min
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2601.19114
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914292819820544
author Chen, Lin
He, Yue
Zhang, Fengting
Wang, Yaonan
Lin, Fengming
Chen, Xiang
Liu, Min
author_facet Chen, Lin
He, Yue
Zhang, Fengting
Wang, Yaonan
Lin, Fengming
Chen, Xiang
Liu, Min
contents Traditional image registration methods are robust but slow due to their iterative nature. While deep learning has accelerated inference, it often struggles with domain shifts. Emerging registration foundation models offer a balance of speed and robustness, yet typically cannot match the peak accuracy of specialized models trained on specific datasets. To mitigate this limitation, we propose Reg-TTR, a test-time refinement framework that synergizes the complementary strengths of both deep learning and conventional registration techniques. By refining the predictions of pre-trained models at inference, our method delivers significantly improved registration accuracy at a modest computational cost, requiring only 21% additional inference time (0.56s). We evaluate Reg-TTR on two distinct tasks and show that it achieves state-of-the-art (SOTA) performance while maintaining inference speeds close to previous deep learning methods. As foundation models continue to emerge, our framework offers an efficient strategy to narrow the performance gap between registration foundation models and SOTA methods trained on specialized datasets. The source code will be publicly available following the acceptance of this work.
format Preprint
id arxiv_https___arxiv_org_abs_2601_19114
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Reg-TTR, Test-Time Refinement for Fast, Robust and Accurate Image Registration
Chen, Lin
He, Yue
Zhang, Fengting
Wang, Yaonan
Lin, Fengming
Chen, Xiang
Liu, Min
Computer Vision and Pattern Recognition
Traditional image registration methods are robust but slow due to their iterative nature. While deep learning has accelerated inference, it often struggles with domain shifts. Emerging registration foundation models offer a balance of speed and robustness, yet typically cannot match the peak accuracy of specialized models trained on specific datasets. To mitigate this limitation, we propose Reg-TTR, a test-time refinement framework that synergizes the complementary strengths of both deep learning and conventional registration techniques. By refining the predictions of pre-trained models at inference, our method delivers significantly improved registration accuracy at a modest computational cost, requiring only 21% additional inference time (0.56s). We evaluate Reg-TTR on two distinct tasks and show that it achieves state-of-the-art (SOTA) performance while maintaining inference speeds close to previous deep learning methods. As foundation models continue to emerge, our framework offers an efficient strategy to narrow the performance gap between registration foundation models and SOTA methods trained on specialized datasets. The source code will be publicly available following the acceptance of this work.
title Reg-TTR, Test-Time Refinement for Fast, Robust and Accurate Image Registration
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.19114