Prior based Sampling for Adaptive LiDAR
This work addresses adaptive sampling for LiDAR systems, offering a domain-specific improvement for autonomous driving and robotics, though it is incremental as it builds on existing depth completion methods.
The authors tackled the problem of adaptive LiDAR sampling by proposing SampleDepth, a CNN that uses prior depth samples to predict sampling masks, optimizing for depth completion tasks. They demonstrated its effectiveness on two datasets and depth completion networks, showing it improves performance in downstream applications.
We propose SampleDepth, a Convolutional Neural Network (CNN), that is suited for an adaptive LiDAR. Typically,LiDAR sampling strategy is pre-defined, constant and independent of the observed scene. Instead of letting a LiDAR sample the scene in this agnostic fashion, SampleDepth determines, adaptively, where it is best to sample the current frame. To do that, SampleDepth uses depth samples from previous time steps to predict a sampling mask for the current frame. Crucially, SampleDepth is trained to optimize the performance of a depth completion downstream task. SampleDepth is evaluated on two different depth completion networks and two LiDAR datasets, KITTI Depth Completion and the newly introduced synthetic dataset, SHIFT. We show that SampleDepth is effective and suitable for different depth completion downstream tasks.