Hierarchical Metric Learning and Matching for 2D and 3D Geometric Correspondences

TL;DR

This paper introduces a hierarchical metric learning approach that leverages multi-level CNN features and activation maps to improve geometric correspondence matching in 2D and 3D, outperforming existing methods.

Contribution

It proposes a novel hierarchical feature supervision and activation map-based matching framework, enhancing descriptor effectiveness for geometric matching tasks.

Findings

Achieves state-of-the-art results on multiple datasets

Demonstrates improved matching accuracy over traditional multi-resolution pyramids

Shows strong generalization across different geometric matching tasks

Abstract

Interest point descriptors have fueled progress on almost every problem in computer vision. Recent advances in deep neural networks have enabled task-specific learned descriptors that outperform hand-crafted descriptors on many problems. We demonstrate that commonly used metric learning approaches do not optimally leverage the feature hierarchies learned in a Convolutional Neural Network (CNN), especially when applied to the task of geometric feature matching. While a metric loss applied to the deepest layer of a CNN, is often expected to yield ideal features irrespective of the task, in fact the growing receptive field as well as striding effects cause shallower features to be better at high precision matching tasks. We leverage this insight together with explicit supervision at multiple levels of the feature hierarchy for better regularization, to learn more effective descriptors in the context of geometric matching tasks. Further, we propose to use activation maps at different layers of a CNN, as an effective and principled replacement for the multi-resolution image pyramids often used for matching tasks. We propose concrete CNN architectures employing these ideas, and evaluate them on multiple datasets for 2D and 3D geometric matching as well as optical flow, demonstrating state-of-the-art results and generalization across datasets.