SConE: Siamese Constellation Embedding Descriptor for Image Matching

TL;DR

This paper introduces SConE, a novel descriptor combining local features and geometric layout of keypoints using a Siamese network, improving image matching accuracy over existing descriptors.

Contribution

It presents a discriminative, low-dimensional descriptor that incorporates geometric constellation information, enhancing matching robustness beyond traditional local feature descriptors.

Findings

Significant performance improvement on TUM dataset.

Descriptor outperforms traditional local feature descriptors.

Effective integration of geometric layout in feature embedding.

Abstract

Numerous computer vision applications rely on local feature descriptors, such as SIFT, SURF or FREAK, for image matching. Although their local character makes image matching processes more robust to occlusions, it often leads to geometrically inconsistent keypoint matches that need to be filtered out, e.g. using RANSAC. In this paper we propose a novel, more discriminative, descriptor that includes not only local feature representation, but also information about the geometric layout of neighbouring keypoints. To that end, we use a Siamese architecture that learns a low-dimensional feature embedding of keypoint constellation by maximizing the distances between non-corresponding pairs of matched image patches, while minimizing it for correct matches. The 48-dimensional oating point descriptor that we train is built on top of the state-of-the-art FREAK descriptor achieves significant performance improvement over the competitors on a challenging TUM dataset.