Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation

Antoine Rimboux; Rob Dupre; Eldriona Daci; Thomas Lagkas; Panagiotis Sarigiannidis; Paolo Remagnino; Vasileios Argyriou

doi:10.1109/DCOSS.2019.00070

Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation

Antoine Rimboux
, Rob Dupre
, Eldriona Daci
, Thomas Lagkas
, Panagiotis Sarigiannidis
, Paolo Remagnino
, Vasileios Argyriou

Kingston University
Link Campus University
University of Sheffield
University of Western Macedonia

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Smart video sensors for applications related to surveillance and security are IOT-based as they use Internet for various purposes. Such applications include crowd behaviour monitoring and advanced decision support systems operating and transmitting information over internet. The analysis of crowd and pedestrian behaviour is an important task for smart IoT cameras and in particular video processing. In order to provide related behavioural models, simulation and tracking approaches have been considered in the literature. In both cases ground truth is essential to train deep models and provide a meaningful quantitative evaluation. We propose a framework for crowd simulation and automatic data generation and annotation that supports multiple cameras and multiple targets. The proposed approach is based on synthetically generated human agents, augmented frames and compositing techniques combined with path finding and planning methods. A number of popular crowd and pedestrian data sets were used to validate the model, and scenarios related to annotation and simulation were considered.

Original language	English
Title of host publication	This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703
DOIs	https://doi.org/10.1109/DCOSS.2019.00070
Publication status	Published - May 2019

Bibliographical note

Note: This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA.

Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703

Keywords

Computer science and informatics

Access to Document

10.1109/DCOSS.2019.00070

https://doi.org/10.1109/DCOSS.2019.00070

Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation
Rimboux, A., Dupre, R., Daci, E., Lagkas, T., Sarigiannidis, P., Remagnino, P. & Argyriou, V., May 2019.
Research output: Contribution to conference › Paper › peer-review

Cite this

Rimboux, A., Dupre, R., Daci, E., Lagkas, T., Sarigiannidis, P., Remagnino, P., & Argyriou, V. (2019). Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation. In This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703 https://doi.org/10.1109/DCOSS.2019.00070

Rimboux, Antoine ; Dupre, Rob ; Daci, Eldriona et al. / Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation. This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703. 2019.

@inproceedings{4139236e8eaa4abf94ff6073300978fc,

title = "Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation",

abstract = "Smart video sensors for applications related to surveillance and security are IOT-based as they use Internet for various purposes. Such applications include crowd behaviour monitoring and advanced decision support systems operating and transmitting information over internet. The analysis of crowd and pedestrian behaviour is an important task for smart IoT cameras and in particular video processing. In order to provide related behavioural models, simulation and tracking approaches have been considered in the literature. In both cases ground truth is essential to train deep models and provide a meaningful quantitative evaluation. We propose a framework for crowd simulation and automatic data generation and annotation that supports multiple cameras and multiple targets. The proposed approach is based on synthetically generated human agents, augmented frames and compositing techniques combined with path finding and planning methods. A number of popular crowd and pedestrian data sets were used to validate the model, and scenarios related to annotation and simulation were considered.",

keywords = "Computer science and informatics",

author = "Antoine Rimboux and Rob Dupre and Eldriona Daci and Thomas Lagkas and Panagiotis Sarigiannidis and Paolo Remagnino and Vasileios Argyriou",

note = "Note: This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703",

year = "2019",

month = may,

doi = "10.1109/DCOSS.2019.00070",

language = "English",

booktitle = "This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703",

}

Rimboux, A, Dupre, R, Daci, E, Lagkas, T, Sarigiannidis, P, Remagnino, P & Argyriou, V 2019, Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation. in This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703. https://doi.org/10.1109/DCOSS.2019.00070

Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation. / Rimboux, Antoine; Dupre, Rob; Daci, Eldriona et al.
This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703. 2019.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation

AU - Rimboux, Antoine

AU - Dupre, Rob

AU - Daci, Eldriona

AU - Lagkas, Thomas

AU - Sarigiannidis, Panagiotis

AU - Remagnino, Paolo

AU - Argyriou, Vasileios

N1 - Note: This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703

PY - 2019/5

Y1 - 2019/5

N2 - Smart video sensors for applications related to surveillance and security are IOT-based as they use Internet for various purposes. Such applications include crowd behaviour monitoring and advanced decision support systems operating and transmitting information over internet. The analysis of crowd and pedestrian behaviour is an important task for smart IoT cameras and in particular video processing. In order to provide related behavioural models, simulation and tracking approaches have been considered in the literature. In both cases ground truth is essential to train deep models and provide a meaningful quantitative evaluation. We propose a framework for crowd simulation and automatic data generation and annotation that supports multiple cameras and multiple targets. The proposed approach is based on synthetically generated human agents, augmented frames and compositing techniques combined with path finding and planning methods. A number of popular crowd and pedestrian data sets were used to validate the model, and scenarios related to annotation and simulation were considered.

AB - Smart video sensors for applications related to surveillance and security are IOT-based as they use Internet for various purposes. Such applications include crowd behaviour monitoring and advanced decision support systems operating and transmitting information over internet. The analysis of crowd and pedestrian behaviour is an important task for smart IoT cameras and in particular video processing. In order to provide related behavioural models, simulation and tracking approaches have been considered in the literature. In both cases ground truth is essential to train deep models and provide a meaningful quantitative evaluation. We propose a framework for crowd simulation and automatic data generation and annotation that supports multiple cameras and multiple targets. The proposed approach is based on synthetically generated human agents, augmented frames and compositing techniques combined with path finding and planning methods. A number of popular crowd and pedestrian data sets were used to validate the model, and scenarios related to annotation and simulation were considered.

KW - Computer science and informatics

U2 - 10.1109/DCOSS.2019.00070

DO - 10.1109/DCOSS.2019.00070

M3 - Conference contribution

BT - This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703

ER -

Rimboux A, Dupre R, Daci E, Lagkas T, Sarigiannidis P, Remagnino P et al. Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation. In This work is co-funded by the NATO within the WITNESS project under grant agreement number G5437. The Titan X Pascal used for this research was donated by NVIDIA. Published in: 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). Institute of Electrical and Electronics Engineers, Inc. ISSN (online) 2325-2944 ISBN 9781728105703. 2019 doi: 10.1109/DCOSS.2019.00070

Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Research output

Smart IoT cameras for crowd analysis based on augmentation for automatic pedestrian detection, simulation and annotation

Cite this