DepthLab: From Partial to Complete

TL;DR

DepthLab is a novel depth inpainting model that effectively handles incomplete depth data, maintaining scale consistency and excelling in various 3D tasks by leveraging image diffusion priors.

Contribution

The paper introduces DepthLab, a depth inpainting approach that is resilient to depth deficiencies and preserves scale, advancing the state-of-the-art in depth completion and related applications.

Findings

Outperforms existing methods in numerical metrics.

Provides high-quality visual reconstructions.

Effective across multiple downstream 3D tasks.

Abstract

Missing values remain a common challenge for depth data across its wide range of applications, stemming from various causes like incomplete data acquisition and perspective alteration. This work bridges this gap with DepthLab, a foundation depth inpainting model powered by image diffusion priors. Our model features two notable strengths: (1) it demonstrates resilience to depth-deficient regions, providing reliable completion for both continuous areas and isolated points, and (2) it faithfully preserves scale consistency with the conditioned known depth when filling in missing values. Drawing on these advantages, our approach proves its worth in various downstream tasks, including 3D scene inpainting, text-to-3D scene generation, sparse-view reconstruction with DUST3R, and LiDAR depth completion, exceeding current solutions in both numerical performance and visual quality. Our project page with source code is available at https://johanan528.github.io/depthlab_web/.