Learning Point Processes and Convolutional Neural Networks for object detection in satellite images - Inria - Institut national de recherche en sciences et technologies du numérique
Pré-Publication, Document De Travail (Working Paper) Année : 2023

Learning Point Processes and Convolutional Neural Networks for object detection in satellite images

Résumé

Convolutional Neural Networks have shown great results for object detection tasks by learning texture and pattern extraction filters. However, object-level interactions are harder to grasp without increasing the complexity of the architectures. On the other hand Point Process models propose to solve the detection on the configuration of objects as a whole, allowing to factor-in the image data, and the objects' interaction priors. In this paper we propose combining the information extracted by a CNN with priors on objects within a Markov Marked Point Process framework. We also demonstrate a method to learn the parameters of this Energy Based Model. We apply this model to the detection of small vehicles in optical satellite imagery, where the image information needs to be complemented with object interaction priors because of noise and small object sizes. The code will be made available at github.com/Ayana-Inria.
Fichier principal
Vignette du fichier
manuscript.pdf (15.35 Mo) Télécharger le fichier
appendices.pdf (148.88 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04250540 , version 1 (19-10-2023)
hal-04250540 , version 2 (14-03-2024)
hal-04250540 , version 3 (15-10-2024)

Licence

Identifiants

  • HAL Id : hal-04250540 , version 1

Citer

Jules Mabon, Mathias Ortner, Josiane Zerubia. Learning Point Processes and Convolutional Neural Networks for object detection in satellite images. 2023. ⟨hal-04250540v1⟩
170 Consultations
49 Téléchargements

Partager

More