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](

Pixelwise Adaptive Discretization with Uncertainty Sampling for Depth Completion

Aggregating Feature Point Cloud for Depth Completion

Self-Paced Knowledge Distillation for Real-Time Image Guided Depth Completion

Least Square Estimation Network for Depth Completion

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

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

Robust Depth Completion with Uncertainty-Driven Loss Functions

A Real-Time Semi-Dense Depth-Guided Depth Completion Network

PENet: Towards Precise and Efficient Image Guided Depth Completion

Progressive Depth Decoupling and Modulating for Flexible Depth Completion

Depth map enhancement based on color and depth consistency

Self-supervised Depth Completion with Attention-Based Loss

Dense Depth Completion Based on Piecewise Planar Model

RigNet++: Efficient Repetitive Image Guided Network for Depth Completion

SDC: Spatial Depth Completion for Outdoor Scenes

Depth Map Completion by Jointly Exploiting Blurry Color Images and Sparse Depth Maps

Depth-Independent Depth Completion via Least Square Estimation

Depth Completion from Sparse LiDAR Data with Depth-Normal Constraints

A Concise but High-performing Network for Image Guided Depth Completion in Autonomous Driving

Adaptive Discrete Disparity Volume for Self-supervised Monocular Depth Estimation