GaussianFlow: Splatting Gaussian Dynamics for 4D Content Creation

• Published in: arXiv.org 2024-05
• Main Authors: Gao, Quankai, Xu, Qiangeng, Cao, Zhe, Mildenhall, Ben, Ma, Wenchao, Chen, Le, Tang, Danhang, Neumann, Ulrich
• Format: Article
• Language: English
• Subjects: Dynamics; Optical flow (image analysis); Synthesis; Video
• ISSN: 2331-8422

Creating 4D fields of Gaussian Splatting from images or videos is a challenging task due to its under-constrained nature. While the optimization can draw photometric reference from the input videos or be regulated by generative models, directly supervising Gaussian motions remains underexplored. In this paper, we introduce a novel concept, Gaussian flow, which connects the dynamics of 3D Gaussians and pixel velocities between consecutive frames. The Gaussian flow can be efficiently obtained by splatting Gaussian dynamics into the image space. This differentiable process enables direct dynamic supervision from optical flow. Our method significantly benefits 4D dynamic content generation and 4D novel view synthesis with Gaussian Splatting, especially for contents with rich motions that are hard to be handled by existing methods. The common color drifting issue that happens in 4D generation is also resolved with improved Guassian dynamics. Superior visual quality on extensive experiments demonstrates our method's effectiveness. Quantitative and qualitative evaluations show that our method achieves state-of-the-art results on both tasks of 4D generation and 4D novel view synthesis. Project page: https://zerg-overmind.github.io/GaussianFlow.github.io/