NM-Net: Mining Reliable Neighbors for Robust Feature Correspondences
This work addresses a domain-specific issue in computer vision for tasks reliant on feature matching, offering an incremental improvement over existing methods.
The paper tackled the problem of unreliable neighbor selection in feature correspondence by proposing a compatibility-specific mining method and a hierarchical network, NM-Net, which achieved state-of-the-art performance on four datasets with varying inlier ratios and feature consistencies.
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.