Generative Autoregressive Ensembles for Satellite Imagery Manipulation Detection
This addresses the issue of tampered satellite images damaging applications like agriculture and disaster assessment, but it appears incremental as it builds on existing generative models for detection.
The paper tackles the problem of detecting unknown manipulations in satellite imagery by using ensembles of generative autoregressive models to model pixel distributions, achieving accurate localization results compared to prior methods.
Satellite imagery is becoming increasingly accessible due to the growing number of orbiting commercial satellites. Many applications make use of such images: agricultural management, meteorological prediction, damage assessment from natural disasters, or cartography are some of the examples. Unfortunately, these images can be easily tampered and modified with image manipulation tools damaging downstream applications. Because the nature of the manipulation applied to the image is typically unknown, unsupervised methods that don't require prior knowledge of the tampering techniques used are preferred. In this paper, we use ensembles of generative autoregressive models to model the distribution of the pixels of the image in order to detect potential manipulations. We evaluate the performance of the presented approach obtaining accurate localization results compared to previously presented approaches.