ANFluid: Animate Natural Fluid Photos base on Physics-Aware Simulation and Dual-Flow Texture Learning
Oct 28, 2024·
,,,,·
1 min read
Xiangcheng Zhai
Yingqi Jie
Xueguang Xie
Aimin Hao
Na Jiang
Yang Gao
Abstract
ANFluid introduces a framework for generating realistic fluid animations from a single static image by combining physics-aware simulation and dual-flow texture learning. Physics-aware simulation encourages motion that follows physical principles, while dual-flow texture learning improves texture prediction through self-supervised training. The framework improves physical consistency and content alignment without increasing model parameters, and supports interactive editing for dynamic content creation.
Type
Publication
In Proceedings of the 32nd ACM International Conference on Multimedia (MM ‘24)
ANFluid focuses on animating natural fluid photos from a single static input. It combines a physics-aware simulation module for physically plausible motion with dual-flow texture learning for improved texture prediction and content alignment.
Citation: Xiangcheng Zhai, Yingqi Jie, Xueguang Xie, Aimin Hao, Na Jiang, and Yang Gao. 2024. ANFluid: Animate Natural Fluid Photos base on Physics-Aware Simulation and Dual-Flow Texture Learning. In Proceedings of the 32nd ACM International Conference on Multimedia (MM ‘24). Association for Computing Machinery, New York, NY, USA, 3323–3331.
Authors
Ph.D. Student in Computer Science
Yingqi Jie is a Ph.D. student at Shanghai Jiao Tong University, advised by Prof. Linpeng Huang and Prof. Sheng’an Zheng. His current work focuses on Linux kernel memory management, memory tiering, NUMA-aware page migration, and evidence-driven systems performance analysis, with prior research experience in microkernel-based operating systems, computer graphics, and AI.