Learning to Generate 4D LiDAR Sequences

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Liang, Ao, Liu, Youquan, Yang, Yu, Lu, Dongyue, Li, Linfeng, Kong, Lingdong, Zhao, Huaici, Ooi, Wei Tsang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912587723046912
author Liang, Ao
Liu, Youquan
Yang, Yu
Lu, Dongyue
Li, Linfeng
Kong, Lingdong
Zhao, Huaici
Ooi, Wei Tsang
author_facet Liang, Ao
Liu, Youquan
Yang, Yu
Lu, Dongyue
Li, Linfeng
Kong, Lingdong
Zhao, Huaici
Ooi, Wei Tsang
contents While generative world models have advanced video and occupancy-based data synthesis, LiDAR generation remains underexplored despite its importance for accurate 3D perception. Extending generation to 4D LiDAR data introduces challenges in controllability, temporal stability, and evaluation. We present LiDARCrafter, a unified framework that converts free-form language into editable LiDAR sequences. Instructions are parsed into ego-centric scene graphs, which a tri-branch diffusion model transforms into object layouts, trajectories, and shapes. A range-image diffusion model generates the initial scan, and an autoregressive module extends it into a temporally coherent sequence. The explicit layout design further supports object-level editing, such as insertion or relocation. To enable fair assessment, we provide EvalSuite, a benchmark spanning scene-, object-, and sequence-level metrics. On nuScenes, LiDARCrafter achieves state-of-the-art fidelity, controllability, and temporal consistency, offering a foundation for LiDAR-based simulation and data augmentation.
format Preprint
id arxiv_https___arxiv_org_abs_2509_11959
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning to Generate 4D LiDAR Sequences
Liang, Ao
Liu, Youquan
Yang, Yu
Lu, Dongyue
Li, Linfeng
Kong, Lingdong
Zhao, Huaici
Ooi, Wei Tsang
Computer Vision and Pattern Recognition
Robotics
While generative world models have advanced video and occupancy-based data synthesis, LiDAR generation remains underexplored despite its importance for accurate 3D perception. Extending generation to 4D LiDAR data introduces challenges in controllability, temporal stability, and evaluation. We present LiDARCrafter, a unified framework that converts free-form language into editable LiDAR sequences. Instructions are parsed into ego-centric scene graphs, which a tri-branch diffusion model transforms into object layouts, trajectories, and shapes. A range-image diffusion model generates the initial scan, and an autoregressive module extends it into a temporally coherent sequence. The explicit layout design further supports object-level editing, such as insertion or relocation. To enable fair assessment, we provide EvalSuite, a benchmark spanning scene-, object-, and sequence-level metrics. On nuScenes, LiDARCrafter achieves state-of-the-art fidelity, controllability, and temporal consistency, offering a foundation for LiDAR-based simulation and data augmentation.
title Learning to Generate 4D LiDAR Sequences
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2509.11959