DepthLab: From Partial to Complete
This work addresses a common challenge in depth data applications, such as 3D reconstruction and scene generation, by providing a robust solution for depth completion, though it appears incremental as it builds on existing diffusion priors.
The paper tackles the problem of missing values in depth data by introducing DepthLab, a foundation depth inpainting model that uses image diffusion priors to complete depth maps, achieving superior numerical performance and visual quality in tasks like 3D scene inpainting and LiDAR depth completion.
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/.