DC Field | Value | Language |
---|---|---|
dc.contributor.author | Na, Ki-In | ko |
dc.contributor.author | Kim, Ue-Hwan | ko |
dc.contributor.author | Kim, Jong-Hwan | ko |
dc.date.accessioned | 2023-07-14T01:00:15Z | - |
dc.date.available | 2023-07-14T01:00:15Z | - |
dc.date.created | 2023-07-13 | - |
dc.date.issued | 2023-08 | - |
dc.identifier.citation | KNOWLEDGE-BASED SYSTEMS, v.274 | - |
dc.identifier.issn | 0950-7051 | - |
dc.identifier.uri | http://hdl.handle.net/10203/310502 | - |
dc.description.abstract | Accurately predicting pedestrian trajectories requires a human-like socio-physical understanding of movement, nearby pedestrians, and obstacles. However, traditional methods struggle to generate multiple trajectories in the same situation based on socio-physical understanding and are compu-tationally intensive, making real-time application difficult. To overcome these limitations, we propose SPU-BERT, a fast multi-trajectory prediction model that incorporates two non-recursive BERTs for multi-goal prediction (MGP) and trajectory-to-goal prediction (TGP). First, MGP predicts multiple goals through generative models, followed by TGP generating trajectories that approach the predicted goals. SPU-BERT can simultaneously understand movement, social interaction, and scene context from trajectories and semantic maps using a single Transformer encoder, providing explainable results as evidence of socio-physical understanding. In experiments, SPU-BERT accurately predicted future trajectories (with 0.19 m and 7.54 pixels of ADE20 for the ETH/UCY datasets and SDD) with over 100 times faster computation (0.132 s) than the state-of-the-art method. The code is available at https://github.com/kina4147/SPUBERT. & COPY; 2023 Published by Elsevier B.V. | - |
dc.language | English | - |
dc.publisher | ELSEVIER | - |
dc.title | SPU-BERT: Faster human multi-trajectory prediction from socio-physical understanding of BERT | - |
dc.type | Article | - |
dc.identifier.wosid | 001013575300001 | - |
dc.identifier.scopusid | 2-s2.0-85162746024 | - |
dc.type.rims | ART | - |
dc.citation.volume | 274 | - |
dc.citation.publicationname | KNOWLEDGE-BASED SYSTEMS | - |
dc.identifier.doi | 10.1016/j.knosys.2023.110637 | - |
dc.contributor.localauthor | Kim, Jong-Hwan | - |
dc.contributor.nonIdAuthor | Na, Ki-In | - |
dc.contributor.nonIdAuthor | Kim, Ue-Hwan | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Pedestrian trajectory prediction | - |
dc.subject.keywordAuthor | Multi-trajectory prediction | - |
dc.subject.keywordAuthor | Socio-physical understanding | - |
dc.subject.keywordAuthor | Transformer | - |
dc.subject.keywordAuthor | BERT | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.