Domain-Shift Immunity in Deep Deformable Registration via Local Feature Representations
Mingzhen Shao, Sarang Joshi · 2025
Read Domain-Shift Immunity in Deep Deformable Registration via Local Feature Representations by Mingzhen Shao, Sarang Joshi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-23142.txt.
Description
Mingzhen Shao and Sarang Joshi challenge a piece of conventional wisdom about deep learning for medical image registration. Learning-based registration, which warps one scan onto another, was assumed fragile under domain shift, and robustness was chased with ever larger training sets. Shao and Joshi argue the opposite: these models are naturally resistant to domain shift because they estimate deformation from local features rather than the image's global appearance.
To isolate that mechanism they built UniReg, which separates feature extraction from deformation estimation, using fixed pretrained extractors feeding a UNet deformation network. Trained on a single dataset, UniReg matched optimization-based methods across domains and imaging modalities. The analysis also pins down why ordinary convolutional models do fail under modality shift: biases absorbed from the training dataset in the early convolutional layers. The design lesson is to protect domain-invariant local features when choosing registration backbones.
Catalog record
| CREATOR | Mingzhen Shao, Sarang Joshi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-23142