Visual Gyroscope: Combination of Deep Learning Features and Direct Alignment for Panoramic Stabilization - Archive ouverte HAL Access content directly
Conference Papers Year : 2023

Visual Gyroscope: Combination of Deep Learning Features and Direct Alignment for Panoramic Stabilization

Abstract

In this article we present a visual gyroscope based on equirectangular panoramas. We propose a new pipeline where we take advantage of combining three different methods to obtain a robust and accurate estimation of the attitude of the camera. We quantitatively and qualitatively validate our method on two image sequences taken with a 360 degrees dual-fisheye camera mounted on different aerial vehicles.
Fichier principal
Vignette du fichier
Berenguel-Baeta_Visual_Gyroscope_Combination_of_Deep_Learning_Features_and_Direct_Alignment_CVPRW_2023_paper.pdf (8.95 Mo) Télécharger le fichier
Origin Publisher files allowed on an open archive

Dates and versions

hal-04066957 , version 1 (04-06-2023)

Identifiers

  • HAL Id : hal-04066957 , version 1

Cite

Bruno Berenguel-Baeta, Antoine André, Guillaume Caron, Jesus Bermudez-Cameo, Josechu Guerrero. Visual Gyroscope: Combination of Deep Learning Features and Direct Alignment for Panoramic Stabilization. IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshop on Omnidirectional Computer Vision, IEEE, Jun 2023, Vancouver, Canada. pp.6444-6447. ⟨hal-04066957⟩
101 View
12 Download

Share

Gmail Mastodon Facebook X LinkedIn More