Location-Velocity Attention for Pedestrian Trajectory Prediction

Research output: Chapter in Book/Conference paperConference paper

Abstract

Pedestrian path forecasting is crucial in applications such as smart video surveillance. It is a challenging task because of the complex crowd movement patterns in the scenes. Most of existing state-of-the-art LSTM based prediction methods require rich context like labelled static obstacles, labelled entrance/exit regions and even the background scene. Furthermore, incorporating contextual information into trajectory prediction increases the computational overhead and decreases the generalization of the prediction models across different scenes. In this paper, we propose a joint Location-Velocity Attention LSTM based method to predict trajectories. Specifically, a module is designed to tweak the LSTM network and an attention mechanism is trained to learn to optimally combine the location
and the velocity information of pedestrians in the prediction process. We have evaluated our approach against other baselines and state-of-the-art methods on several publicly available datasets. The results show that it not only outperforms other prediction methods but it also has a good generalization ability.
Original languageEnglish
Title of host publication2019 IEEE Winter Conference on Applications of Computer Vision
Place of PublicationUSA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages2038–2047
Number of pages10
ISBN (Electronic)9781728119755
DOIs
Publication statusPublished - 4 Mar 2019
Event19th IEEE Winter Conference on Applications of Computer Vision, WACV 2019 - Waikoloa Village, United States
Duration: 7 Jan 201911 Jan 2019

Publication series

NameProceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019

Conference

Conference19th IEEE Winter Conference on Applications of Computer Vision, WACV 2019
Abbreviated titleVACV 2019
CountryUnited States
CityWaikoloa Village
Period7/01/1911/01/19

Fingerprint

Trajectories

Cite this

Xue, H., Huynh, D., & Reynolds, M. (2019). Location-Velocity Attention for Pedestrian Trajectory Prediction. In 2019 IEEE Winter Conference on Applications of Computer Vision (pp. 2038–2047). [8659060] (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019). USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/WACV.2019.00221
Xue, Hao ; Huynh, Du ; Reynolds, Mark. / Location-Velocity Attention for Pedestrian Trajectory Prediction. 2019 IEEE Winter Conference on Applications of Computer Vision. USA : IEEE, Institute of Electrical and Electronics Engineers, 2019. pp. 2038–2047 (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019).
@inproceedings{e6eb1da088ae4a789fc831919333f178,
title = "Location-Velocity Attention for Pedestrian Trajectory Prediction",
abstract = "Pedestrian path forecasting is crucial in applications such as smart video surveillance. It is a challenging task because of the complex crowd movement patterns in the scenes. Most of existing state-of-the-art LSTM based prediction methods require rich context like labelled static obstacles, labelled entrance/exit regions and even the background scene. Furthermore, incorporating contextual information into trajectory prediction increases the computational overhead and decreases the generalization of the prediction models across different scenes. In this paper, we propose a joint Location-Velocity Attention LSTM based method to predict trajectories. Specifically, a module is designed to tweak the LSTM network and an attention mechanism is trained to learn to optimally combine the location and the velocity information of pedestrians in the prediction process. We have evaluated our approach against other baselines and state-of-the-art methods on several publicly available datasets. The results show that it not only outperforms other prediction methods but it also has a good generalization ability.",
author = "Hao Xue and Du Huynh and Mark Reynolds",
year = "2019",
month = "3",
day = "4",
doi = "10.1109/WACV.2019.00221",
language = "English",
series = "Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
pages = "2038–2047",
booktitle = "2019 IEEE Winter Conference on Applications of Computer Vision",
address = "United States",

}

Xue, H, Huynh, D & Reynolds, M 2019, Location-Velocity Attention for Pedestrian Trajectory Prediction. in 2019 IEEE Winter Conference on Applications of Computer Vision., 8659060, Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019, IEEE, Institute of Electrical and Electronics Engineers, USA, pp. 2038–2047, 19th IEEE Winter Conference on Applications of Computer Vision, WACV 2019, Waikoloa Village, United States, 7/01/19. https://doi.org/10.1109/WACV.2019.00221

Location-Velocity Attention for Pedestrian Trajectory Prediction. / Xue, Hao; Huynh, Du; Reynolds, Mark.

2019 IEEE Winter Conference on Applications of Computer Vision. USA : IEEE, Institute of Electrical and Electronics Engineers, 2019. p. 2038–2047 8659060 (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019).

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Location-Velocity Attention for Pedestrian Trajectory Prediction

AU - Xue, Hao

AU - Huynh, Du

AU - Reynolds, Mark

PY - 2019/3/4

Y1 - 2019/3/4

N2 - Pedestrian path forecasting is crucial in applications such as smart video surveillance. It is a challenging task because of the complex crowd movement patterns in the scenes. Most of existing state-of-the-art LSTM based prediction methods require rich context like labelled static obstacles, labelled entrance/exit regions and even the background scene. Furthermore, incorporating contextual information into trajectory prediction increases the computational overhead and decreases the generalization of the prediction models across different scenes. In this paper, we propose a joint Location-Velocity Attention LSTM based method to predict trajectories. Specifically, a module is designed to tweak the LSTM network and an attention mechanism is trained to learn to optimally combine the location and the velocity information of pedestrians in the prediction process. We have evaluated our approach against other baselines and state-of-the-art methods on several publicly available datasets. The results show that it not only outperforms other prediction methods but it also has a good generalization ability.

AB - Pedestrian path forecasting is crucial in applications such as smart video surveillance. It is a challenging task because of the complex crowd movement patterns in the scenes. Most of existing state-of-the-art LSTM based prediction methods require rich context like labelled static obstacles, labelled entrance/exit regions and even the background scene. Furthermore, incorporating contextual information into trajectory prediction increases the computational overhead and decreases the generalization of the prediction models across different scenes. In this paper, we propose a joint Location-Velocity Attention LSTM based method to predict trajectories. Specifically, a module is designed to tweak the LSTM network and an attention mechanism is trained to learn to optimally combine the location and the velocity information of pedestrians in the prediction process. We have evaluated our approach against other baselines and state-of-the-art methods on several publicly available datasets. The results show that it not only outperforms other prediction methods but it also has a good generalization ability.

U2 - 10.1109/WACV.2019.00221

DO - 10.1109/WACV.2019.00221

M3 - Conference paper

T3 - Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019

SP - 2038

EP - 2047

BT - 2019 IEEE Winter Conference on Applications of Computer Vision

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - USA

ER -

Xue H, Huynh D, Reynolds M. Location-Velocity Attention for Pedestrian Trajectory Prediction. In 2019 IEEE Winter Conference on Applications of Computer Vision. USA: IEEE, Institute of Electrical and Electronics Engineers. 2019. p. 2038–2047. 8659060. (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019). https://doi.org/10.1109/WACV.2019.00221