Adaptive LiDAR Sampling and Depth Completion using Ensemble Variance

TL;DR

This paper introduces an adaptive sampling method for depth completion that uses ensemble variance to select pixels, significantly reducing the number of measurements needed while maintaining high accuracy in autonomous vehicle applications.

Contribution

The work presents a novel ensemble variance-based adaptive sampling technique for depth completion, compatible with any black-box depth predictor, and demonstrates substantial measurement reduction.

Findings

Achieves state-of-the-art accuracy on KITTI dataset.

Reduces required measurements by 4-10 times compared to random sampling.

Effective with neural networks and Random Forests.

Abstract

This work considers the problem of depth completion, with or without image data, where an algorithm may measure the depth of a prescribed limited number of pixels. The algorithmic challenge is to choose pixel positions strategically and dynamically to maximally reduce overall depth estimation error. This setting is realized in daytime or nighttime depth completion for autonomous vehicles with a programmable LiDAR. Our method uses an ensemble of predictors to define a sampling probability over pixels. This probability is proportional to the variance of the predictions of ensemble members, thus highlighting pixels that are difficult to predict. By additionally proceeding in several prediction phases, we effectively reduce redundant sampling of similar pixels. Our ensemble-based method may be implemented using any depth-completion learning algorithm, such as a state-of-the-art neural network, treated as a black box. In particular, we also present a simple and effective Random Forest-based algorithm, and similarly use its internal ensemble in our design. We conduct experiments on the KITTI dataset, using the neural network algorithm of Ma et al. and our Random Forest based learner for implementing our method. The accuracy of both implementations exceeds the state of the art. Compared with a random or grid sampling pattern, our method allows a reduction by a factor of 4-10 in the number of measurements required to attain the same accuracy.