Abstract:Image-guided depth completion aims at generating a dense depth map from sparse LiDAR data and RGB image. Recent methods have shown promising performance by reformulating it as a classification problem with two sub-tasks: depth discretization and probability prediction. They divide the depth range into several discrete depth values as depth categories, serving as priors for scene depth distributions. However, previous depth discretization methods are easy to be impacted by depth distribution variations across different scenes, resulting in suboptimal scene depth distribution priors. To address the above problem, we propose a progressive depth decoupling and modulating network, which incrementally decouples the depth range into bins and adaptively generates multi-scale dense depth maps in multiple stages. Specifically, we first design a Bins Initializing Module (BIM) to construct the seed bins by exploring the depth distribution information within a sparse depth map, adapting variations of depth distribution. Then, we devise an incremental depth decoupling branch to progressively refine the depth distribution information from global to local. Meanwhile, an adaptive depth modulating branch is developed to progressively improve the probability representation from coarse-grained to fine-grained. And the bi-directional information interactions are proposed to strengthen the information interaction between those two branches (sub-tasks) for promoting information complementation in each branch. Further, we introduce a multi-scale supervision mechanism to learn the depth distribution information in latent features and enhance the adaptation capability across different scenes. Experimental results on public datasets demonstrate that our method outperforms the state-of-the-art methods. The code will be open-sourced at [this https URL](

OMNI-DC: Highly Robust Depth Completion with Multiresolution Depth Integration

OGNI-DC: Robust Depth Completion with Optimization-Guided Neural Iterations

SparseDC: Depth Completion from sparse and non-uniform inputs

Aggregating Feature Point Cloud for Depth Completion

MFF-Net: Towards Efficient Monocular Depth Completion With Multi-Modal Feature Fusion

Depth Coefficients for Depth Completion

DCDepth: Progressive Monocular Depth Estimation in Discrete Cosine Domain

Marigold-DC: Zero-Shot Monocular Depth Completion with Guided Diffusion

Depth Completion from Sparse LiDAR Data with Depth-Normal Constraints

GeoDC: Geometry-Constrained Depth Completion With Depth Distribution Modeling

RigNet++: Semantic Assisted Repetitive Image Guided Network for Depth Completion

Depth Completion Towards Different Sensor Configurations Via Relative Depth Map Estimation and Scale Recovery

Adaptive Context-Aware Multi-Modal Network for Depth Completion

Towards Domain-agnostic Depth Completion

Progressive Depth Decoupling and Modulating for Flexible Depth Completion

SDC: Spatial Depth Completion for Outdoor Scenes

A Compromise Principle in Deep Monocular Depth Estimation

Deep Depth Completion from Extremely Sparse Data: A Survey

Robust Depth Completion with Uncertainty-Driven Loss Functions

A Sparsity-Invariant Model Via Unifying Depth Prediction and Completion

Towards Comprehensive Monocular Depth Estimation: Multiple Heads are Better Than One