Details
| Original language | English |
|---|---|
| Title of host publication | 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) |
| Publisher | IEEE Computer Society |
| Pages | 2039-2049 |
| Number of pages | 11 |
| ISBN (electronic) | 9798350365474 |
| ISBN (print) | 979-8-3503-6548-1 |
| Publication status | Published - 17 Jun 2024 |
| Event | 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2024 - Seattle, United States Duration: 16 Jun 2024 → 22 Jun 2024 |
Publication series
| Name | IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops |
|---|---|
| ISSN (Print) | 2160-7508 |
| ISSN (electronic) | 2160-7516 |
Abstract
Existing trajectory prediction methods for autonomous driving typically rely on one-stage trajectory prediction models, which condition future trajectories on observed trajectories combined with fused scene information. However, they often struggle with complex scene constraints, such as those encountered at intersections. To this end, we present a novel method, called LAformer. It uses an attention-based temporally dense lane-aware estimation module to continuously estimate the likelihood of the alignment between motion dynamics and scene information extracted from an HD map. Additionally, unlike one-stage prediction models, LAformer utilizes predictions from the first stage as anchor trajectories. It leverages a second-stage motion refinement module to further explore temporal consistency across the complete time horizon. Extensive experiments on nuScenes and Argoverse 1 demonstrate that LAformer achieves excellent generalized performance for multimodal trajectory prediction. The source code of LAformer is available at https://github.com/mengmengliu1998/LAformer.
Keywords
- lane-aware selection, motion refinement, multimodal, Trajectory prediction
ASJC Scopus subject areas
- Computer Science(all)
- Computer Vision and Pattern Recognition
- Engineering(all)
- Electrical and Electronic Engineering
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). IEEE Computer Society, 2024. p. 2039-2049 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - LAformer
T2 - 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2024
AU - Liu, Mengmeng
AU - Cheng, Hao
AU - Chen, Lin
AU - Broszio, Hellward
AU - Li, Jiangtao
AU - Zhao, Runjiang
AU - Sester, Monika
AU - Yang, Michael Ying
N1 - Publisher Copyright: © 2024 IEEE.
PY - 2024/6/17
Y1 - 2024/6/17
N2 - Existing trajectory prediction methods for autonomous driving typically rely on one-stage trajectory prediction models, which condition future trajectories on observed trajectories combined with fused scene information. However, they often struggle with complex scene constraints, such as those encountered at intersections. To this end, we present a novel method, called LAformer. It uses an attention-based temporally dense lane-aware estimation module to continuously estimate the likelihood of the alignment between motion dynamics and scene information extracted from an HD map. Additionally, unlike one-stage prediction models, LAformer utilizes predictions from the first stage as anchor trajectories. It leverages a second-stage motion refinement module to further explore temporal consistency across the complete time horizon. Extensive experiments on nuScenes and Argoverse 1 demonstrate that LAformer achieves excellent generalized performance for multimodal trajectory prediction. The source code of LAformer is available at https://github.com/mengmengliu1998/LAformer.
AB - Existing trajectory prediction methods for autonomous driving typically rely on one-stage trajectory prediction models, which condition future trajectories on observed trajectories combined with fused scene information. However, they often struggle with complex scene constraints, such as those encountered at intersections. To this end, we present a novel method, called LAformer. It uses an attention-based temporally dense lane-aware estimation module to continuously estimate the likelihood of the alignment between motion dynamics and scene information extracted from an HD map. Additionally, unlike one-stage prediction models, LAformer utilizes predictions from the first stage as anchor trajectories. It leverages a second-stage motion refinement module to further explore temporal consistency across the complete time horizon. Extensive experiments on nuScenes and Argoverse 1 demonstrate that LAformer achieves excellent generalized performance for multimodal trajectory prediction. The source code of LAformer is available at https://github.com/mengmengliu1998/LAformer.
KW - lane-aware selection
KW - motion refinement
KW - multimodal
KW - Trajectory prediction
UR - http://www.scopus.com/inward/record.url?scp=85206383198&partnerID=8YFLogxK
U2 - 10.48550/arXiv.2302.13933
DO - 10.48550/arXiv.2302.13933
M3 - Conference contribution
SN - 979-8-3503-6548-1
T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
SP - 2039
EP - 2049
BT - 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
PB - IEEE Computer Society
Y2 - 16 June 2024 through 22 June 2024
ER -