D2D: Keypoint Extraction with Describe to Detect Approach

TL;DR

This paper introduces D2D, a novel keypoint extraction method that inverts traditional strategies by first describing image regions and then detecting keypoints based on descriptor information, enhancing matching performance.

Contribution

The paper proposes a descriptor-based keypoint detection approach that requires no additional training and improves performance across multiple benchmarks.

Findings

Enhances image matching accuracy

Generalizes across different descriptors and tasks

Improves keypoint detection without extra training

Abstract

In this paper, we present a novel approach that exploits the information within the descriptor space to propose keypoint locations. Detect then describe, or detect and describe jointly are two typical strategies for extracting local descriptors. In contrast, we propose an approach that inverts this process by first describing and then detecting the keypoint locations. % Describe-to-Detect (D2D) leverages successful descriptor models without the need for any additional training. Our method selects keypoints as salient locations with high information content which is defined by the descriptors rather than some independent operators. We perform experiments on multiple benchmarks including image matching, camera localisation, and 3D reconstruction. The results indicate that our method improves the matching performance of various descriptors and that it generalises across methods and tasks.