$S^3$: Learnable Sparse Signal Superdensity for Guided Depth Estimation
This work addresses a domain-specific problem for applications such as robotics, 3D reconstruction, and augmented reality, offering an incremental improvement by enhancing existing guided depth estimation methods.
The paper tackles the problem of limited improvement in dense depth estimation from sparse signals like LiDAR and Radar due to low density and imbalanced distribution, proposing the $S^3$ technique to expand depth values and estimate confidence, with experiments showing effectiveness, robustness, and flexibility on LiDAR and Radar signals.
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.