@inproceedings{Tian:2021:10.1007/978-3-030-69535-4_14,
author = {Tian, Y and Balntas, V and Ng, T and Barroso-Laguna, A and Demiris, Y and Mikolajczyk, K},
doi = {10.1007/978-3-030-69535-4_14},
pages = {223--240},
title = {D2D: Keypoint Extraction with Describe to Detect Approach},
url = {http://dx.doi.org/10.1007/978-3-030-69535-4_14},
year = {2021}
}

Download

TY  - CPAPER
AB  - In this paper, we present a novel approach that exploits the information within the descriptor space to propose keypoint locations. Detect then describe, or jointly detect and describe are two typical strategies for extracting local features. 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 are 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.
AU  - Tian,Y
AU  - Balntas,V
AU  - Ng,T
AU  - Barroso-Laguna,A
AU  - Demiris,Y
AU  - Mikolajczyk,K
DO  - 10.1007/978-3-030-69535-4_14
EP  - 240
PY  - 2021///
SN  - 0302-9743
SP  - 223
TI  - D2D: Keypoint Extraction with Describe to Detect Approach
UR  - http://dx.doi.org/10.1007/978-3-030-69535-4_14
ER  - 

Download