# RIFT2: Speeding-up RIFT with A New Rotation-Invariance Technique

arXiv: https://arxiv.org/abs/2303.00319 · HTML: https://mlsanity.com/paper/2303.00319/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 1 of its 24 references that are in the map.
- First posted on arXiv 1 Mar 2023.

## Builds on (most cited first, 1 of 1)

- [Robust Registration of Multimodal Remote Sensing Images Based on Structural Similarity](https://mlsanity.com/paper/2103.16871/) (2021)

## Built on by (most cited first, 11 of 11)

- [3MOS: Multi-sources, Multi-resolutions, and Multi-scenes dataset for Optical-SAR image matching](https://mlsanity.com/paper/2404.00838/) (2024)
- [MIFNet: Learning Modality-Invariant Features for Generalizable Multimodal Image Matching](https://mlsanity.com/paper/2501.11299/) (2025)
- [MapGlue: Multimodal Remote Sensing Image Matching](https://mlsanity.com/paper/2503.16185/) (2025)
- [GDROS: A Geometry-Guided Dense Registration Framework for Optical-SAR Images under Large Geometric Transformations](https://mlsanity.com/paper/2511.00598/) (2025)
- [CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration](https://mlsanity.com/paper/2604.05689/) (2026)
- [XPoint: A Self-Supervised Visual-State-Space based Architecture for Multispectral Image Registration](https://mlsanity.com/paper/2411.07430/) (2024)
- [Comparative Evaluation of Traditional and Deep Learning Feature Matching Algorithms using Chandrayaan-2 Lunar Data](https://mlsanity.com/paper/2509.04775/) (2025)
- [Matching of SAR and optical images based on transformation to shared modality](https://mlsanity.com/paper/2602.12515/) (2026)
- [RBE-Flow: Recurrent Bayesian Estimation on Feature Manifolds for Cross-Modal Registration](https://mlsanity.com/paper/2606.30492/) (2026)
- [Self-Calibrating Dense Displacement Fields for Reliable Co-Registration of Large Optical Satellite Imagery](https://mlsanity.com/paper/2608.22300/) (2026)
- [Multimodal Remote Sensing Image Registration: A Comprehensive Review, Challenges and Prospects](https://mlsanity.com/paper/2610.11176/) (2026)

## Abstract (arXiv)

Multimodal image matching is an important prerequisite for multisource image information fusion. Compared with the traditional matching problem, multimodal feature matching is more challenging due to the severe nonlinear radiation distortion (NRD). Radiation-variation insensitive feature transform (RIFT)~citeli2019rift has shown very good robustness to NRD and become a baseline method in multimodal feature matching. However, the high computational cost for rotation invariance largely limits its usage in practice. In this paper, we propose an improved RIFT method, called RIFT2. We develop a new rotation invariance technique based on dominant index value, which avoids the construction process of convolution sequence ring. Hence, it can speed up the running time and reduce the memory consumption of the original RIFT by almost 3 times in theory. Extensive experiments show that RIFT2 achieves similar matching performance to RIFT while being much faster and having less memory consumption. The source code will be made publicly available in urlhttps://github.com/LJY-RS/RIFT2-multimodal-matching-rotation