Skip to Main content Skip to Navigation
Conference papers

A comparative study of semantic segmentation using omnidirectional images

Ahmed Rida Sekkat 1, 2 Yohan Dupuis 3 Paul Honeine 1 Pascal Vasseur 2
1 DocApp - LITIS - Equipe Apprentissage
LITIS - Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes
2 STI - LITIS - Equipe Systèmes de Transport Intelligent
LITIS - Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes
Abstract : The semantic segmentation of omnidirectional urban driving images is a research topic that has increasingly attracted the attention of researchers. This paper presents a thorough comparative study of different neural network models trained on four different representations: perspective, equirectangular, spherical and fisheye. We use in this study real perspective images, and synthetic perspective, fisheye and equirectangular images, as well as a test set of real fisheye images. We evaluate the performance of convolution on spherical images and perspective images. The conclusions obtained by analyzing the results of this study are multiple and help understanding how different networks learn to deal with omnidirectional distortions. Our main finding is that models trained on omnidirectional images are robust against modality changes and are able to learn a universal representation, giving good results in both perspective and omnidirectional images. The relevance of all results is examined with an analysis of quantitative measures.
Complete list of metadatas

https://hal-normandie-univ.archives-ouvertes.fr/hal-03088368
Contributor : Paul Honeine <>
Submitted on : Saturday, December 26, 2020 - 10:35:46 AM
Last modification on : Wednesday, January 13, 2021 - 3:38:26 AM

File

RFIAP_2020_paper_47.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03088368, version 1

Citation

Ahmed Rida Sekkat, Yohan Dupuis, Paul Honeine, Pascal Vasseur. A comparative study of semantic segmentation using omnidirectional images. Congrès Reconnaissance des Formes, Image, Apprentissage et Perception (RFIAP), Jun 2020, Vannes, France. ⟨hal-03088368⟩

Share

Metrics

Record views

15

Files downloads

7