# Geometry-Aware Sparse Depth Sampling for High-Fidelity RGB-D Depth Completion in Robotic Systems

arXiv: https://arxiv.org/abs/2512.08229 · HTML: https://mlsanity.com/paper/2512.08229/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 39 references that are in the map.
- First posted on arXiv 9 Dec 2025.

## Builds on (most cited first, 18 of 18)

- [SuperPoint: Self-Supervised Interest Point Detection and Description](https://mlsanity.com/paper/1712.07629/) (2017)
- [SceneGraphFusion: Incremental 3D Scene Graph Prediction from RGB-D Sequences](https://mlsanity.com/paper/2103.14898/) (2021)
- [Deep Depth Completion of a Single RGB-D Image](https://mlsanity.com/paper/1803.09326/) (2018)
- [RigNet: Repetitive Image Guided Network for Depth Completion](https://mlsanity.com/paper/2107.13802/) (2021)
- [Marigold: Affordable Adaptation of Diffusion-Based Image Generators for Image Analysis](https://mlsanity.com/paper/2505.09358/) (2025)
- [RGB-D Local Implicit Function for Depth Completion of Transparent Objects](https://mlsanity.com/paper/2104.00622/) (2021)
- [Bilateral Propagation Network for Depth Completion](https://mlsanity.com/paper/2403.11270/) (2024)
- [Marigold-DC: Zero-Shot Monocular Depth Completion with Guided Diffusion](https://mlsanity.com/paper/2412.13389/) (2024)
- [Deep Depth Completion from Extremely Sparse Data: A Survey](https://mlsanity.com/paper/2205.05335/) (2022)
- [RigNet++: Semantic Assisted Repetitive Image Guided Network for Depth Completion](https://mlsanity.com/paper/2309.00655/) (2023)
- [ProtoDepth: Unsupervised Continual Depth Completion with Prototypes](https://mlsanity.com/paper/2503.12745/) (2025)
- [SparseDC: Depth Completion from sparse and non-uniform inputs](https://mlsanity.com/paper/2312.00097/) (2023)
- [Learning an Efficient Multimodal Depth Completion Model](https://mlsanity.com/paper/2208.10771/) (2022)
- [Masked Spatial Propagation Network for Sparsity-Adaptive Depth Refinement](https://mlsanity.com/paper/2404.19294/) (2024)
- [URNet: Uncertainty-aware Refinement Network for Event-based Stereo Depth Estimation](https://mlsanity.com/paper/2509.18184/) (2025)
- [3D Instance Segmentation Using Deep Learning on RGB-D Indoor Data](https://mlsanity.com/paper/2406.14581/) (2024)
- [DenseFormer: Learning Dense Depth Map from Sparse Depth and Image via Conditional Diffusion Model](https://mlsanity.com/paper/2503.23993/) (2025)
- [Improving Optical Flow and Stereo Depth Estimation by Leveraging Uncertainty-Based Learning Difficulties](https://mlsanity.com/paper/2506.00324/) (2025)

## Abstract (arXiv)

Accurate three-dimensional perception is essential for modern industrial robotic systems that perform manipulation, inspection, and navigation tasks. RGB-D and stereo vision sensors are widely used for this purpose, but the depth maps they produce are often noisy, incomplete, or biased due to sensor limitations and environmental conditions. Depth completion methods aim to generate dense, reliable depth maps from RGB images and sparse depth input. However, a key limitation in current depth completion pipelines is the unrealistic generation of sparse depth: sparse pixels are typically selected uniformly at random from dense ground-truth depth, ignoring the fact that real sensors exhibit geometry-dependent and spatially nonuniform reliability. In this work, we propose a normal-guided sparse depth sampling strategy that leverages PCA-based surface normal estimation on the RGB-D point cloud to compute a per-pixel depth reliability measure. The sparse depth samples are then drawn according to this reliability distribution. We integrate this sampling method with the Marigold-DC diffusion-based depth completion model and evaluate it on NYU Depth v2 using the standard metrics. Experiments show that our geometry-aware sparse depth improves accuracy, reduces artifacts near edges and discontinuities, and produces more realistic training conditions that better reflect real sensor behavior.