Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training

Abstract In the area of pedestrian trajectory prediction, the hybrid structures of temporal feature extractor or spatial feature extractor have paved the way for the precise prediction model, and they are in larger and larger scale. Learning of specific feature encoding model not only influenced by...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Zonggui Yao, Jun Yu, Jiajun Ding
Formato:	article
Lenguaje:	EN
Publicado:	Wiley 2021
Materias:	Photography TR1-1050 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/a221bb83f854462eb70822cb29fb44d5
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a221bb83f854462eb70822cb29fb44d5
record_format	dspace
spelling	oai:doaj.org-article:a221bb83f854462eb70822cb29fb44d52021-11-29T03:38:16ZContrastive learning of graph encoder for accelerating pedestrian trajectory prediction training1751-96671751-965910.1049/ipr2.12185https://doaj.org/article/a221bb83f854462eb70822cb29fb44d52021-12-01T00:00:00Zhttps://doi.org/10.1049/ipr2.12185https://doaj.org/toc/1751-9659https://doaj.org/toc/1751-9667Abstract In the area of pedestrian trajectory prediction, the hybrid structures of temporal feature extractor or spatial feature extractor have paved the way for the precise prediction model, and they are in larger and larger scale. Learning of specific feature encoding model not only influenced by the structure of the network, but also by the learning manners such as supervised learning and unsupervised learning. Previous works concentrated on more comprehensive encoders and more delicate designs of feature extractors. However, the mutual influence factors from the neighbour pedestrians associate with the distance to the centre pedestrian seldomly noticed. Most of the existed feature extractors in prediction models trained in the way of supervised learning other than unsupervised manners caused the problem that the extracted features are always handcrafted without the natural distinction of obscure situations. The graph contrastive accelerating encoder is proposed, which accelerates the pedestrian trajectory prediction training process of the state of the art method of spatio‐temporal graph transformer networks. Employing the unsupervised contrastive learning process and the graph of neighbours representing distance affection of nearest and farthest pedestrian to the centre pedestrian, the graph contrastive accelerating encoder significantly shrinked the training time. Holding the final performance on to state of the art level, the proposed method let the lowest pedestrian trajectory prediction error show up in the obviously earlier training steps.Zonggui YaoJun YuJiajun DingWileyarticlePhotographyTR1-1050Computer softwareQA76.75-76.765ENIET Image Processing, Vol 15, Iss 14, Pp 3645-3660 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Photography TR1-1050 Computer software QA76.75-76.765
spellingShingle	Photography TR1-1050 Computer software QA76.75-76.765 Zonggui Yao Jun Yu Jiajun Ding Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training
description	Abstract In the area of pedestrian trajectory prediction, the hybrid structures of temporal feature extractor or spatial feature extractor have paved the way for the precise prediction model, and they are in larger and larger scale. Learning of specific feature encoding model not only influenced by the structure of the network, but also by the learning manners such as supervised learning and unsupervised learning. Previous works concentrated on more comprehensive encoders and more delicate designs of feature extractors. However, the mutual influence factors from the neighbour pedestrians associate with the distance to the centre pedestrian seldomly noticed. Most of the existed feature extractors in prediction models trained in the way of supervised learning other than unsupervised manners caused the problem that the extracted features are always handcrafted without the natural distinction of obscure situations. The graph contrastive accelerating encoder is proposed, which accelerates the pedestrian trajectory prediction training process of the state of the art method of spatio‐temporal graph transformer networks. Employing the unsupervised contrastive learning process and the graph of neighbours representing distance affection of nearest and farthest pedestrian to the centre pedestrian, the graph contrastive accelerating encoder significantly shrinked the training time. Holding the final performance on to state of the art level, the proposed method let the lowest pedestrian trajectory prediction error show up in the obviously earlier training steps.
format	article
author	Zonggui Yao Jun Yu Jiajun Ding
author_facet	Zonggui Yao Jun Yu Jiajun Ding
author_sort	Zonggui Yao
title	Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training
title_short	Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training
title_full	Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training
title_fullStr	Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training
title_full_unstemmed	Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training
title_sort	contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training
publisher	Wiley
publishDate	2021
url	https://doaj.org/article/a221bb83f854462eb70822cb29fb44d5
work_keys_str_mv	AT zongguiyao contrastivelearningofgraphencoderforacceleratingpedestriantrajectorypredictiontraining AT junyu contrastivelearningofgraphencoderforacceleratingpedestriantrajectorypredictiontraining AT jiajunding contrastivelearningofgraphencoderforacceleratingpedestriantrajectorypredictiontraining
_version_	1718407664644390912

Contrastive learning of graph encoder for accelerating pedestrian trajectory prediction training

Ejemplares similares