BAGS: Building Animatable Gaussian Splatting from a Monocular Video with Diffusion Priors
Preprint
Abstract
Animatable 3D reconstruction has significant applications across various fields, primarily relying on artists' handcraft creation. Recently, some studies have successfully constructed animatable 3D models from monocular videos. However, these approaches require sufficient view coverage of the object within the input video and typically necessitate significant time and computational costs for training and rendering. This limitation restricts the practical applications. In this work, we propose a method to build animatable 3D Gaussian Splatting from monocular video with diffusion priors. The 3D Gaussian representations significantly accelerate the training and rendering process, and the diffusion priors allow the method to learn 3D models with limited viewpoints. We also present the rigid regularization to enhance the utilization of the priors. We perform an extensive evaluation across various real-world videos, demonstrating its superior performance compared to the current state-of-the-art methods.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- DreaMo: Articulated 3D Reconstruction from a Single Casual Video2025
- ReconFusion: 3D Reconstruction with Diffusion Priors2024
- GART: Gaussian Articulated Template Models2024
- DreamBooth3D: Subject-Driven Text-to-3D Generation2023
- Neural Novel Actor: Learning a Generalized Animatable Neural Representation for Human ActorsIEEE Transactions on Visualization and Computer Graphics, 2023
- DOVE: Learning Deformable 3D Objects by Watching VideosInternational Journal of Computer Vision, 2023
- NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis2020
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric2018
- Non-rigid structure from motion with complementary rank-3 spaces2011
- Particle Video: Long-Range Motion Estimation Using Point TrajectoriesInternational Journal of Computer Vision, 2008