RangeLDM: Fast Realistic LiDAR Point Cloud Generation
This addresses the high cost and scaling challenges of physical LiDAR sensors for autonomous driving, though it appears incremental as it builds on existing generative models.
The authors tackled the problem of slow and unrealistic LiDAR point cloud generation for autonomous driving by introducing RangeLDM, which uses latent diffusion models to achieve fast and high-quality generation, demonstrating robust expressiveness and speed on KITTI-360 and nuScenes datasets.
Autonomous driving demands high-quality LiDAR data, yet the cost of physical LiDAR sensors presents a significant scaling-up challenge. While recent efforts have explored deep generative models to address this issue, they often consume substantial computational resources with slow generation speeds while suffering from a lack of realism. To address these limitations, we introduce RangeLDM, a novel approach for rapidly generating high-quality range-view LiDAR point clouds via latent diffusion models. We achieve this by correcting range-view data distribution for accurate projection from point clouds to range images via Hough voting, which has a critical impact on generative learning. We then compress the range images into a latent space with a variational autoencoder, and leverage a diffusion model to enhance expressivity. Additionally, we instruct the model to preserve 3D structural fidelity by devising a range-guided discriminator. Experimental results on KITTI-360 and nuScenes datasets demonstrate both the robust expressiveness and fast speed of our LiDAR point cloud generation.