Depth completion (DC) aims to predict a dense depth map from an RGB image and a sparse depth map. Existing DC methods generalize poorly to new datasets or unseen sparse depth patterns, limiting their real-world applications. We propose OMNI-DC, a highly robust DC model that generalizes well zero-shot to various datasets. The key design is a novel Multi-resolution Depth Integrator, allowing our model to deal with very sparse depth inputs. We also introduce a novel Laplacian loss to model the ambiguity in the training process. Moreover, we train OMNI-DC on a mixture of high-quality datasets with a scale normalization technique and synthetic depth patterns. Extensive experiments on 7 datasets show consistent improvements over baselines, reducing errors by as much as 43%. Codes and checkpoints are available at https://github.com/princeton-vl/OMNI-DC.
@article{arxiv.2411.19278,
title = {OMNI-DC: Highly Robust Depth Completion with Multiresolution Depth Integration},
author = {Yiming Zuo and Willow Yang and Zeyu Ma and Jia Deng},
journal= {arXiv preprint arXiv:2411.19278},
year = {2025}
}
Comments
Accepted to ICCV 2025. Added additional results and ablations