Attention-Based Learning for Fluid State Interpolation and Editing in a Time-Continuous Framework
This addresses the problem of generating smooth and sharp fluid animations for computer graphics applications, representing an incremental improvement in domain-specific methods.
The paper tackles fluid animation interpolation by introducing FluidsFormer, a transformer-based method that combines PITT and RNN to predict fluid properties in continuous time, enabling substep frame interpolation between simulated keyframes. It demonstrates improved temporal smoothness and sharpness in smoke interpolation, with initial experiments on liquids.
In this work, we introduce FluidsFormer: a transformer-based approach for fluid interpolation within a continuous-time framework. By combining the capabilities of PITT and a residual neural network (RNN), we analytically predict the physical properties of the fluid state. This enables us to interpolate substep frames between simulated keyframes, enhancing the temporal smoothness and sharpness of animations. We demonstrate promising results for smoke interpolation and conduct initial experiments on liquids.