a Multi-Perspective Approach to Anomaly Detection for Self -Aware Embodied Agents
Mohamad Baydoun, Mahdyar Ravanbakhsh, Damian Campo, Pablo Marin, David Martín, Lucio Marcenaro, Andrea Cavallaro, Carlo S. Regazzoni
Abstract
This paper focuses on multi-sensor anomaly detection for moving cognitive agents using both external and private first-person visual observations. Both observation types are used to characterize agents motion in a given environment. The proposed method generates locally uniform motion models by dividing a Gaussian process that approximates agents displacements on the scene and provides a Shared Level (SL) self-awareness based on Environment Centered (EC) models. Such models are then used to train in a semi-unsupervised way a set of Generative Adversarial Networks (GANs) that produce an estimation of external and internal parameters of moving agents. Obtained results exemplify the feasibility of using multi-perspective data for predicting and analyzing trajectory information.
BibTeX
@inproceedings{icassp2018_amultiperspectiv,
title = {a Multi-Perspective Approach to Anomaly Detection for Self -Aware Embodied Agents},
author = {Mohamad Baydoun and Mahdyar Ravanbakhsh and Damian Campo and Pablo Marin and David Martín and Lucio Marcenaro and Andrea Cavallaro and Carlo S. Regazzoni},
booktitle = {ICASSP 2018},
year = {2018}
}