We tackle the efficiency problem of learning local feature matching. Recent advancements have given rise to purely CNN-based and transformer-based approaches, each augmented with deep learning techniques. While CNN-based methods often excel in matching speed, transformer-based methods tend to provide more accurate matches. We propose an efficient transformer-based network architecture for local feature matching. This technique is built on constructing multiple homography hypotheses to approximate the continuous correspondence in the real world and uni-directional cross-attention to accelerate the refinement. On the YFCC100M dataset, our matching accuracy is competitive with LoFTR, a state-of-the-art transformer-based architecture, while the inference speed is boosted to 4 times, even outperforming the CNN-based methods. Comprehensive evaluations on other open datasets such as Megadepth, ScanNet, and HPatches demonstrate our method's efficacy, highlighting its potential to significantly enhance a wide array of downstream applications.
@article{arxiv.2410.22733,
title = {ETO:Efficient Transformer-based Local Feature Matching by Organizing Multiple Homography Hypotheses},
author = {Junjie Ni and Guofeng Zhang and Guanglin Li and Yijin Li and Xinyang Liu and Zhaoyang Huang and Hujun Bao},
journal= {arXiv preprint arXiv:2410.22733},
year = {2025}
}