# RoTIR: Rotation-Equivariant Network and Transformers for Fish Scale Image Registration

arXiv: https://arxiv.org/abs/2401.11270 · HTML: https://mlsanity.com/paper/2401.11270/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jan 2024.

## Built on by (most cited first, 2 of 2)

- [Moment kernels: a simple and scalable approach for equivariance to rotations and reflections in deep convolutional networks](https://mlsanity.com/paper/2505.21736/) (2025)
- [Rotation Equivariant Convolutions in Deformable Registration of Brain MRI](https://mlsanity.com/paper/2604.08034/) (2026)

## Abstract (arXiv)

Image registration is an essential process for aligning features of interest from multiple images. With the recent development of deep learning techniques, image registration approaches have advanced to a new level. In this work, we present 'Rotation-Equivariant network and Transformers for Image Registration' (RoTIR), a deep-learning-based method for the alignment of fish scale images captured by light microscopy. This approach overcomes the challenge of arbitrary rotation and translation detection, as well as the absence of ground truth data. We employ feature-matching approaches based on Transformers and general E(2)-equivariant steerable CNNs for model creation. Besides, an artificial training dataset is employed for semi-supervised learning. Results show RoTIR successfully achieves the goal of fish scale image registration.