Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Learning to Generate Realistic LiDAR Point Clouds

About

We present LiDARGen, a novel, effective, and controllable generative model that produces realistic LiDAR point cloud sensory readings. Our method leverages the powerful score-matching energy-based model and formulates the point cloud generation process as a stochastic denoising process in the equirectangular view. This model allows us to sample diverse and high-quality point cloud samples with guaranteed physical feasibility and controllability. We validate the effectiveness of our method on the challenging KITTI-360 and NuScenes datasets. The quantitative and qualitative results show that our approach produces more realistic samples than other generative models. Furthermore, LiDARGen can sample point clouds conditioned on inputs without retraining. We demonstrate that our proposed generative model could be directly used to densify LiDAR point clouds. Our code is available at: https://www.zyrianov.org/lidargen/

Vlas Zyrianov, Xiyue Zhu, Shenlong Wang• 2022

Related benchmarks

TaskDatasetResultRank
LiDAR Semantic SegmentationSemanticKITTI
mIoU60.39
36
Unconditional LiDAR GenerationKITTI360 (val)
FSVD39.2
11
LiDAR Scene GenerationKITTI-360 (val)
FRD579.4
9
LiDAR DensificationKITTI-360 64-beam, ~120K to ~250K (val)
CD (m)0.2948
9
LiDAR DensificationnuScenes 32-beam (val)
CD (m)0.3287
9
Unconditional LiDAR GenerationKITTI-360 19
FRD579.4
8
LiDAR Scene GenerationnuScenes 2
FPD22.8
7
LiDAR GenerationnuScenes (val)
MMD19
6
LiDAR GenerationnuScenes v1.0-trainval (val)
MMD19
6
LiDAR Scene GenerationKITTI-360 (sequences 0-1)
MMD_BEV4.80e-4
5
Showing 10 of 11 rows

Other info

Follow for update