# Transport-Based Neural Style Transfer for Smoke Simulations

## TL;DR

This paper introduces a novel transport-based neural style transfer method for volumetric smoke simulations, enabling content-aware and physically consistent stylizations from natural images to complex smoke flows.

## Contribution

It presents the first transport-based neural style transfer algorithm for 3D smoke data, allowing detailed and semantically rich style transfer with physical plausibility.

## Key findings

- Enables style transfer from natural images to smoke simulations.
- Ensures temporal consistency in stylized smoke sequences.
- Supports multi-view 3D reconstructions of stylized smoke.

## Abstract

Artistically controlling fluids has always been a challenging task. Optimization techniques rely on approximating simulation states towards target velocity or density field configurations, which are often handcrafted by artists to indirectly control smoke dynamics. Patch synthesis techniques transfer image textures or simulation features to a target flow field. However, these are either limited to adding structural patterns or augmenting coarse flows with turbulent structures, and hence cannot capture the full spectrum of different styles and semantically complex structures. In this paper, we propose the first Transport-based Neural Style Transfer (TNST) algorithm for volumetric smoke data. Our method is able to transfer features from natural images to smoke simulations, enabling general content-aware manipulations ranging from simple patterns to intricate motifs. The proposed algorithm is physically inspired, since it computes the density transport from a source input smoke to a desired target configuration. Our transport-based approach allows direct control over the divergence of the stylization velocity field by optimizing incompressible and irrotational potentials that transport smoke towards stylization. Temporal consistency is ensured by transporting and aligning subsequent stylized velocities, and 3D reconstructions are computed by seamlessly merging stylizations from different camera viewpoints.