Neural Kinematic Bases for Fluids
This work addresses the need for efficient and physically accurate fluid simulations in graphics and animation, though it appears incremental as it builds on existing neural methods for fluid dynamics.
The paper tackles the problem of real-time fluid simulation by proposing mesh-free simulations using kinematic neural bases for velocity fields, achieving real-time animation with properties like orthogonality and divergence-free flow.
We propose mesh-free fluid simulations that exploit a kinematic neural basis for velocity fields represented by an MLP. We design a set of losses that ensures that these neural bases approximate fundamental physical properties such as orthogonality, divergence-free, boundary alignment, and smoothness. Our neural bases can then be used to fit an input sketch of a flow, which will inherit the same fundamental properties from the bases. We then can animate such flow in real-time using standard time integrators. Our neural bases can accommodate different domains, moving boundaries, and naturally extend to three dimensions.