DeformSplat: Rigidity-Aware 3D Gaussian Deformation from a Single Image
Siggraph ASIA 2025
TL;DR: Given a single target image and an initial 3D Gaussian, DeformSplat deforms the Gaussian to match the target image while preserving geometry. The motion is represented by varying the transparency of the images over time.
Abstract
Reconstructing object deformation from a single image remains a significant challenge in computer vision and graphics. Existing methods typically rely on multi-view video to recover deformation, limiting their applicability under constrained scenarios. To address this, we propose DeformSplat, a novel framework that effectively guides 3D Gaussian deformation from only a single image. Our method introduces two main technical contributions. First, we present a Gaussian-to-Pixel matching that bridges the domain gap between 3D Gaussian representations and 2D pixel observations. This enables robust deformation guidance from sparse visual cues. Second, we propose Rigid Part Segmentation consisting of initialization and refinement. This segmentation explicitly identifies rigid regions, crucial for maintaining geometric coherence during deformation. By combining these two techniques, our approach can reconstruct consistent deformations from a single image. Extensive experiments demonstrate that our approach significantly outperforms existing methods and naturally extends to various applications, such as frame interpolation and interactive object manipulation.
Methods
DeformSplat first establishes Gaussian-to-Pixel Matching that connects the 3D Gaussian representation with the 2D target image to guide deformation. On the other hand, Rigid Part Segmentation method preserves the original geometry by detecting rigid regions and constraining their deformation. By combining two methods, our approach reconstructs the deformed Gaussian, which can be rendered from novel views.
Rigid Part Segmentation
Motion Interpolation
Application: Interactive Manipulation
Application: Multi-frame Interpolation
@misc{kim2025deformsplat,
title={Rigidity-Aware 3D Gaussian Deformation from a Single Image},
author={Jinhyeok Kim and Jaehun Bang and Seunghyun Seo and Kyungdon Joo},
year={2025},
eprint={2509.22222},
archivePrefix={arXiv},
primaryClass={cs.GR},
url={https://arxiv.org/abs/2509.22222},
}