$S^3$: Learnable Sparse Signal Superdensity for Guided Depth Estimation

TL;DR

This paper introduces $S^3$, a learnable method to enhance guided depth estimation by expanding sparse signals like LiDAR and Radar, improving accuracy and robustness across various applications.

Contribution

The paper proposes a novel $S^3$ technique that effectively expands sparse signals and estimates confidence, applicable at multiple stages in depth estimation models.

Findings

$S^3$ improves depth estimation accuracy with sparse signals.

The method is robust across different sparse data sources.

$S^3$ is flexible and can be integrated end-to-end.

Abstract

Dense depth estimation plays a key role in multiple applications such as robotics, 3D reconstruction, and augmented reality. While sparse signal, e.g., LiDAR and Radar, has been leveraged as guidance for enhancing dense depth estimation, the improvement is limited due to its low density and imbalanced distribution. To maximize the utility from the sparse source, we propose $S^3$ technique, which expands the depth value from sparse cues while estimating the confidence of expanded region. The proposed $S^3$ can be applied to various guided depth estimation approaches and trained end-to-end at different stages, including input, cost volume and output. Extensive experiments demonstrate the effectiveness, robustness, and flexibility of the $S^3$ technique on LiDAR and Radar signal.