Pedestrian Trajectory Prediction Using A Social Pyramid

Research output: Chapter in Book/Conference paperConference paper

Abstract

Understanding and forecasting human movement paths are vital for a wide range of real world applications. It is not an easy task to generate plausible future paths as the scenes and human movement patterns are often very complex. In this paper, we propose a social pyramid based prediction method (SPP), which includes two encoders to capture motion and social information. Specifically, we design a social pyramid map structure for the Social encoder, which can differentiate the influence of other pedestrians in nearby areas or remote areas based on their spatial locations. For the Motion encoder, a mixing attention mechanism is proposed to combine the location coordinates and velocity vectors. The two encoded features are then merged and passed to the decoder which generates future paths of pedestrians. Our extensive experimental results demonstrate competitive prediction performance from our method compared to state-of-art methods.
Original languageEnglish
Title of host publicationPRICAI 2019: Trends in Artificial Intelligence
EditorsAbhaya C. Nayak, Alok Sharma
PublisherSpringer International Publishing
Pages439-453
Number of pages14
Volume11671
ISBN (Electronic)9783030299118
ISBN (Print)9783030299101
DOIs
Publication statusPublished - 2019
Event16th Pacific Rim International Conference on Artificial Intelligence 2019 - Cuvu, Fiji
Duration: 26 Aug 201930 Aug 2019

Publication series

NameLecture Notes in Computer Science

Conference

Conference16th Pacific Rim International Conference on Artificial Intelligence 2019
Abbreviated titlePRICAI 2019
CountryFiji
CityCuvu
Period26/08/1930/08/19

Fingerprint

Trajectories

Cite this

Xue, H., Huynh, D., & Reynolds, M. (2019). Pedestrian Trajectory Prediction Using A Social Pyramid. In A. C. Nayak, & A. Sharma (Eds.), PRICAI 2019: Trends in Artificial Intelligence (Vol. 11671, pp. 439-453). (Lecture Notes in Computer Science). Springer International Publishing. https://doi.org/10.1007/978-3-030-29911-8_34
Xue, Hao ; Huynh, Du ; Reynolds, Mark. / Pedestrian Trajectory Prediction Using A Social Pyramid. PRICAI 2019: Trends in Artificial Intelligence. editor / Abhaya C. Nayak ; Alok Sharma. Vol. 11671 Springer International Publishing, 2019. pp. 439-453 (Lecture Notes in Computer Science).
@inproceedings{6e207b6ec34b4aba84dc14e3e10c269c,
title = "Pedestrian Trajectory Prediction Using A Social Pyramid",
abstract = "Understanding and forecasting human movement paths are vital for a wide range of real world applications. It is not an easy task to generate plausible future paths as the scenes and human movement patterns are often very complex. In this paper, we propose a social pyramid based prediction method (SPP), which includes two encoders to capture motion and social information. Specifically, we design a social pyramid map structure for the Social encoder, which can differentiate the influence of other pedestrians in nearby areas or remote areas based on their spatial locations. For the Motion encoder, a mixing attention mechanism is proposed to combine the location coordinates and velocity vectors. The two encoded features are then merged and passed to the decoder which generates future paths of pedestrians. Our extensive experimental results demonstrate competitive prediction performance from our method compared to state-of-art methods.",
author = "Hao Xue and Du Huynh and Mark Reynolds",
year = "2019",
doi = "10.1007/978-3-030-29911-8_34",
language = "English",
isbn = "9783030299101",
volume = "11671",
series = "Lecture Notes in Computer Science",
publisher = "Springer International Publishing",
pages = "439--453",
editor = "Nayak, {Abhaya C.} and Alok Sharma",
booktitle = "PRICAI 2019: Trends in Artificial Intelligence",

}

Xue, H, Huynh, D & Reynolds, M 2019, Pedestrian Trajectory Prediction Using A Social Pyramid. in AC Nayak & A Sharma (eds), PRICAI 2019: Trends in Artificial Intelligence. vol. 11671, Lecture Notes in Computer Science, Springer International Publishing, pp. 439-453, 16th Pacific Rim International Conference on Artificial Intelligence 2019, Cuvu, Fiji, 26/08/19. https://doi.org/10.1007/978-3-030-29911-8_34

Pedestrian Trajectory Prediction Using A Social Pyramid. / Xue, Hao; Huynh, Du; Reynolds, Mark.

PRICAI 2019: Trends in Artificial Intelligence. ed. / Abhaya C. Nayak; Alok Sharma. Vol. 11671 Springer International Publishing, 2019. p. 439-453 (Lecture Notes in Computer Science).

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Pedestrian Trajectory Prediction Using A Social Pyramid

AU - Xue, Hao

AU - Huynh, Du

AU - Reynolds, Mark

PY - 2019

Y1 - 2019

N2 - Understanding and forecasting human movement paths are vital for a wide range of real world applications. It is not an easy task to generate plausible future paths as the scenes and human movement patterns are often very complex. In this paper, we propose a social pyramid based prediction method (SPP), which includes two encoders to capture motion and social information. Specifically, we design a social pyramid map structure for the Social encoder, which can differentiate the influence of other pedestrians in nearby areas or remote areas based on their spatial locations. For the Motion encoder, a mixing attention mechanism is proposed to combine the location coordinates and velocity vectors. The two encoded features are then merged and passed to the decoder which generates future paths of pedestrians. Our extensive experimental results demonstrate competitive prediction performance from our method compared to state-of-art methods.

AB - Understanding and forecasting human movement paths are vital for a wide range of real world applications. It is not an easy task to generate plausible future paths as the scenes and human movement patterns are often very complex. In this paper, we propose a social pyramid based prediction method (SPP), which includes two encoders to capture motion and social information. Specifically, we design a social pyramid map structure for the Social encoder, which can differentiate the influence of other pedestrians in nearby areas or remote areas based on their spatial locations. For the Motion encoder, a mixing attention mechanism is proposed to combine the location coordinates and velocity vectors. The two encoded features are then merged and passed to the decoder which generates future paths of pedestrians. Our extensive experimental results demonstrate competitive prediction performance from our method compared to state-of-art methods.

UR - https://link.springer.com/book/10.1007/978-3-030-29911-8#about

U2 - 10.1007/978-3-030-29911-8_34

DO - 10.1007/978-3-030-29911-8_34

M3 - Conference paper

SN - 9783030299101

VL - 11671

T3 - Lecture Notes in Computer Science

SP - 439

EP - 453

BT - PRICAI 2019: Trends in Artificial Intelligence

A2 - Nayak, Abhaya C.

A2 - Sharma, Alok

PB - Springer International Publishing

ER -

Xue H, Huynh D, Reynolds M. Pedestrian Trajectory Prediction Using A Social Pyramid. In Nayak AC, Sharma A, editors, PRICAI 2019: Trends in Artificial Intelligence. Vol. 11671. Springer International Publishing. 2019. p. 439-453. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-030-29911-8_34