Variational Autoencoder with Gaussian Random Field prior: application to unsupervised animal detection in aerial images

Hugo Gangloff (1) , Minh-Tan Pham (2) , Luc Courtrai (2) , Sébastien Lefèvre (2)

1 MIA Paris-Saclay - Mathématiques et Informatique Appliquées
2 OBELIX - Observation de l’environnement par imagerie complexe

In real world datasets of aerial images, the objects of interest are often missing, hard to annotate and of varying aspects. The framework of unsupervised Anomaly Detection (AD) is highly relevant in this context, and Variational Autoencoders (VAEs), a family of popular probabilistic models, are often used. We develop on the literature of VAEs for AD in order to take advantage of the particular textures that appear in natural aerial images. More precisely we propose a new VAE model with a Gaussian Random Field (GRF) prior (VAE-GRF), which generalize the classical VAE model, and we provide the necessary procedures and hypotheses required for the model to be tractable. We show that, under some assumptions, the VAE-GRF largely outperforms the traditional VAE and some other probabilistic models developed for AD. Our results suggest that the VAE-GRF could be used as a relevant VAE baseline in place of the traditional VAE with very limited additional computational cost. We provide competitive results on the MVTec dataset and two other datasets dedicated to the task of unsupervised animal detection in aerial images.

Mots clés: Gaussian random fields; aerial images; Variational autoencoders; anomaly detection

Domaines: Machine Learning [stat.ML]; Traitement du signal et de l'image [eess.SP]

Hugo Gangloff, Minh-Tan Pham, Luc Courtrai, Sébastien Lefèvre. Variational Autoencoder with Gaussian Random Field prior: application to unsupervised animal detection in aerial images. 2023 IEEE International Conference on Image Processing (ICIP), Oct 2023, Kuala Lumpur, Malaysia. pp.1620-1624, ⟨10.1109/ICIP49359.2023.10222900⟩. ⟨hal-03774853⟩