3DGS-Drag: Dragging Gaussians for Intuitive Point-Based 3D Editing
Authors:
Preprint
Abstract
The transformative potential of 3D content creation has been progressively unlocked through advancements in generative models. Recently, intuitive drag editing with geometric changes has attracted significant attention in 2D editing yet remains challenging for 3D scenes. In this paper, we introduce 3DGS-Drag -- a point-based 3D editing framework that provides efficient, intuitive drag manipulation of real 3D scenes. Our approach bridges the gap between deformation-based and 2D-editing-based 3D editing methods, addressing their limitations to geometry-related content editing. We leverage two key innovations: deformation guidance utilizing 3D Gaussian Splatting for consistent geometric modifications and diffusion guidance for content correction and visual quality enhancement. A progressive editing strategy further supports aggressive 3D drag edits. Our method enables a wide range of edits, including motion change, shape adjustment, inpainting, and content extension. Experimental results demonstrate the effectiveness of 3DGS-Drag in various scenes, achieving state-of-the-art performance in geometry-related 3D content editing. Notably, the editing is efficient, taking 10 to 20 minutes on a single RTX 4090 GPU.Keywords
All Related Versions
This publication has 33 references indexed in Scilit:
- RegionDrag: Fast Region-Based Image Editing with Diffusion Models2024
- Imagic: Text-Based Real Image Editing with Diffusion Models2023
- NeuralEditor: Editing Neural Radiance Fields via Manipulating Point Clouds2023
- Deforming Radiance Fields with Cages2022
- ARF: Artistic Radiance Fields2022
- NeRF-Editing: Geometry Editing of Neural Radiance Fields2022
- High-Resolution Image Synthesis with Latent Diffusion Models2022
- Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields2022
- StylizedNeRF: Consistent 3D Scene Stylization as Stylized NeRF via 2D-3D Mutual Learning2022
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric2018