English

NM-Net: Mining Reliable Neighbors for Robust Feature Correspondences

Computer Vision and Pattern Recognition 2019-04-02 v1

Abstract

Feature correspondence selection is pivotal to many feature-matching based tasks in computer vision. Searching for spatially k-nearest neighbors is a common strategy for extracting local information in many previous works. However, there is no guarantee that the spatially k-nearest neighbors of correspondences are consistent because the spatial distribution of false correspondences is often irregular. To address this issue, we present a compatibility-specific mining method to search for consistent neighbors. Moreover, in order to extract and aggregate more reliable features from neighbors, we propose a hierarchical network named NM-Net with a series of convolution layers taking the generated graph as input, which is insensitive to the order of correspondences. Our experimental results have shown the proposed method achieves the state-of-the-art performance on four datasets with various inlier ratios and varying numbers of feature consistencies.

Keywords

Cite

@article{arxiv.1904.00320,
  title  = {NM-Net: Mining Reliable Neighbors for Robust Feature Correspondences},
  author = {Chen Zhao and Zhiguo Cao and Chi Li and Xin Li and Jiaqi Yang},
  journal= {arXiv preprint arXiv:1904.00320},
  year   = {2019}
}

Comments

IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2019) (oral)

R2 v1 2026-06-23T08:24:14.403Z