LEGO-Motion: Learning-Enhanced Grids with Occupancy Instance Modeling for Class-Agnostic Motion Prediction

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Qian, Kangan, Miao, Jinyu, Luo, Ziang, Fu, Zheng, Li, and Jinchen, Shi, Yining, Wang, Yunlong, Jiang, Kun, Yang, Mengmeng, Yang, Diange
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910867454427136
author Qian, Kangan
Miao, Jinyu
Luo, Ziang
Fu, Zheng
Li, and Jinchen
Shi, Yining
Wang, Yunlong
Jiang, Kun
Yang, Mengmeng
Yang, Diange
author_facet Qian, Kangan
Miao, Jinyu
Luo, Ziang
Fu, Zheng
Li, and Jinchen
Shi, Yining
Wang, Yunlong
Jiang, Kun
Yang, Mengmeng
Yang, Diange
contents Accurate and reliable spatial and motion information plays a pivotal role in autonomous driving systems. However, object-level perception models struggle with handling open scenario categories and lack precise intrinsic geometry. On the other hand, occupancy-based class-agnostic methods excel in representing scenes but fail to ensure physics consistency and ignore the importance of interactions between traffic participants, hindering the model's ability to learn accurate and reliable motion. In this paper, we introduce a novel occupancy-instance modeling framework for class-agnostic motion prediction tasks, named LEGO-Motion, which incorporates instance features into Bird's Eye View (BEV) space. Our model comprises (1) a BEV encoder, (2) an Interaction-Augmented Instance Encoder, and (3) an Instance-Enhanced BEV Encoder, improving both interaction relationships and physics consistency within the model, thereby ensuring a more accurate and robust understanding of the environment. Extensive experiments on the nuScenes dataset demonstrate that our method achieves state-of-the-art performance, outperforming existing approaches. Furthermore, the effectiveness of our framework is validated on the advanced FMCW LiDAR benchmark, showcasing its practical applicability and generalization capabilities. The code will be made publicly available to facilitate further research.
format Preprint
id arxiv_https___arxiv_org_abs_2503_07367
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LEGO-Motion: Learning-Enhanced Grids with Occupancy Instance Modeling for Class-Agnostic Motion Prediction
Qian, Kangan
Miao, Jinyu
Luo, Ziang
Fu, Zheng
Li, and Jinchen
Shi, Yining
Wang, Yunlong
Jiang, Kun
Yang, Mengmeng
Yang, Diange
Computer Vision and Pattern Recognition
Accurate and reliable spatial and motion information plays a pivotal role in autonomous driving systems. However, object-level perception models struggle with handling open scenario categories and lack precise intrinsic geometry. On the other hand, occupancy-based class-agnostic methods excel in representing scenes but fail to ensure physics consistency and ignore the importance of interactions between traffic participants, hindering the model's ability to learn accurate and reliable motion. In this paper, we introduce a novel occupancy-instance modeling framework for class-agnostic motion prediction tasks, named LEGO-Motion, which incorporates instance features into Bird's Eye View (BEV) space. Our model comprises (1) a BEV encoder, (2) an Interaction-Augmented Instance Encoder, and (3) an Instance-Enhanced BEV Encoder, improving both interaction relationships and physics consistency within the model, thereby ensuring a more accurate and robust understanding of the environment. Extensive experiments on the nuScenes dataset demonstrate that our method achieves state-of-the-art performance, outperforming existing approaches. Furthermore, the effectiveness of our framework is validated on the advanced FMCW LiDAR benchmark, showcasing its practical applicability and generalization capabilities. The code will be made publicly available to facilitate further research.
title LEGO-Motion: Learning-Enhanced Grids with Occupancy Instance Modeling for Class-Agnostic Motion Prediction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.07367