Deep Generative Modeling of LiDAR Data
This work addresses a key challenge in robotics by enabling better generative modeling of lidar data, which is incremental as it adapts existing methods to a new domain.
The paper tackles the problem of generating lidar scans, which are crucial for robot mapping and localization, by adapting deep generative models to produce high-quality samples and learn meaningful latent representations, demonstrating significant improvements over state-of-the-art point cloud generation methods.
Building models capable of generating structured output is a key challenge for AI and robotics. While generative models have been explored on many types of data, little work has been done on synthesizing lidar scans, which play a key role in robot mapping and localization. In this work, we show that one can adapt deep generative models for this task by unravelling lidar scans into a 2D point map. Our approach can generate high quality samples, while simultaneously learning a meaningful latent representation of the data. We demonstrate significant improvements against state-of-the-art point cloud generation methods. Furthermore, we propose a novel data representation that augments the 2D signal with absolute positional information. We show that this helps robustness to noisy and imputed input; the learned model can recover the underlying lidar scan from seemingly uninformative data