Saved in:
Bibliographic Details
Main Authors: Barbarani, Giovanni, Vaccarino, Francesco, Trivigno, Gabriele, Guerra, Marco, Berton, Gabriele, Masone, Carlo
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.01315
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914820943511552
author Barbarani, Giovanni
Vaccarino, Francesco
Trivigno, Gabriele
Guerra, Marco
Berton, Gabriele
Masone, Carlo
author_facet Barbarani, Giovanni
Vaccarino, Francesco
Trivigno, Gabriele
Guerra, Marco
Berton, Gabriele
Masone, Carlo
contents In computer vision, keypoint detection is a fundamental task, with applications spanning from robotics to image retrieval; however, existing learning-based methods suffer from scale dependency and lack flexibility. This paper introduces a novel approach that leverages Morse theory and persistent homology, powerful tools rooted in algebraic topology. We propose a novel loss function based on the recent introduction of a notion of subgradient in persistent homology, paving the way toward topological learning. Our detector, MorseDet, is the first topology-based learning model for feature detection, which achieves competitive performance in keypoint repeatability and introduces a principled and theoretically robust approach to the problem.
format Preprint
id arxiv_https___arxiv_org_abs_2406_01315
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Scale-Free Image Keypoints Using Differentiable Persistent Homology
Barbarani, Giovanni
Vaccarino, Francesco
Trivigno, Gabriele
Guerra, Marco
Berton, Gabriele
Masone, Carlo
Computer Vision and Pattern Recognition
Machine Learning
Algebraic Topology
55N31
I.2.10
In computer vision, keypoint detection is a fundamental task, with applications spanning from robotics to image retrieval; however, existing learning-based methods suffer from scale dependency and lack flexibility. This paper introduces a novel approach that leverages Morse theory and persistent homology, powerful tools rooted in algebraic topology. We propose a novel loss function based on the recent introduction of a notion of subgradient in persistent homology, paving the way toward topological learning. Our detector, MorseDet, is the first topology-based learning model for feature detection, which achieves competitive performance in keypoint repeatability and introduces a principled and theoretically robust approach to the problem.
title Scale-Free Image Keypoints Using Differentiable Persistent Homology
topic Computer Vision and Pattern Recognition
Machine Learning
Algebraic Topology
55N31
I.2.10
url https://arxiv.org/abs/2406.01315