Improving Transformer-based Image Matching by Cascaded Capturing Spatially Informative Keypoints

Understand

Learning robust local image feature matching is a fundamental low-level vision task, which has been widely explored in the past few years.

• Recently, detector-free local feature matchers based on transformers have shown promising results, which largely outperform pure Convolutional Neural Network (CNN) based ones.
• But correlations produced by transformer-based methods are spatially limited to the center of source views' coarse patches, because of the costly attention learning.
• In this work, we rethink this issue and find that such matching formulation degrades pose estimation, especially for low-resolution images.