Lidar Upsampling with Sliced Wasserstein Distance
This addresses the problem of sparse lidar data for autonomous driving systems, representing an incremental improvement in sensor domain adaptation.
The paper tackles lidar point cloud upsampling for autonomous driving perception by proposing a method using edge-aware dense convolutions and Sliced Wasserstein Distance, which achieves better upsampling results in experiments.
Lidar became an important component of the perception systems in autonomous driving. But challenges of training data acquisition and annotation made emphasized the role of the sensor to sensor domain adaptation. In this work, we address the problem of lidar upsampling. Learning on lidar point clouds is rather a challenging task due to their irregular and sparse structure. Here we propose a method for lidar point cloud upsampling which can reconstruct fine-grained lidar scan patterns. The key idea is to utilize edge-aware dense convolutions for both feature extraction and feature expansion. Additionally applying a more accurate Sliced Wasserstein Distance facilitates learning of the fine lidar sweep structures. This in turn enables our method to employ a one-stage upsampling paradigm without the need for coarse and fine reconstruction. We conduct several experiments to evaluate our method and demonstrate that it provides better upsampling.