3D Gaussian Inpainting with Depth-Guided Cross-View Consistency
This addresses the challenge of cross-view consistency in 3D inpainting for applications in computer vision and graphics, representing an incremental improvement over existing methods.
The paper tackles the problem of achieving texture and geometry consistency across camera views in 3D inpainting using methods like Neural Radiance Field or 3D Gaussian Splatting, and proposes a framework called 3DGIC that uses depth-guided cross-view consistency to refine inpainting masks, resulting in outperforming current state-of-the-art methods on benchmark datasets.
When performing 3D inpainting using novel-view rendering methods like Neural Radiance Field (NeRF) or 3D Gaussian Splatting (3DGS), how to achieve texture and geometry consistency across camera views has been a challenge. In this paper, we propose a framework of 3D Gaussian Inpainting with Depth-Guided Cross-View Consistency (3DGIC) for cross-view consistent 3D inpainting. Guided by the rendered depth information from each training view, our 3DGIC exploits background pixels visible across different views for updating the inpainting mask, allowing us to refine the 3DGS for inpainting purposes.Through extensive experiments on benchmark datasets, we confirm that our 3DGIC outperforms current state-of-the-art 3D inpainting methods quantitatively and qualitatively.