LidarDM: Generative LiDAR Simulation in a Generated World
This work addresses the need for high-quality LiDAR simulation to train and test perception models in autonomous driving, representing a novel method for a known bottleneck.
The paper tackles the problem of generating realistic and temporally coherent LiDAR videos for autonomous driving simulations, achieving superior performance in realism, temporal coherency, and layout consistency compared to competing algorithms.
We present LidarDM, a novel LiDAR generative model capable of producing realistic, layout-aware, physically plausible, and temporally coherent LiDAR videos. LidarDM stands out with two unprecedented capabilities in LiDAR generative modeling: (i) LiDAR generation guided by driving scenarios, offering significant potential for autonomous driving simulations, and (ii) 4D LiDAR point cloud generation, enabling the creation of realistic and temporally coherent sequences. At the heart of our model is a novel integrated 4D world generation framework. Specifically, we employ latent diffusion models to generate the 3D scene, combine it with dynamic actors to form the underlying 4D world, and subsequently produce realistic sensory observations within this virtual environment. Our experiments indicate that our approach outperforms competing algorithms in realism, temporal coherency, and layout consistency. We additionally show that LidarDM can be used as a generative world model simulator for training and testing perception models.