ScalarFlow: A Large-Scale Volumetric Data Set of Real-world Scalar Transport Flows for Computer Animation and Machine Learning

TL;DR

ScalarFlow is a large-scale dataset of real-world smoke plumes with detailed volumetric reconstructions, enabling advanced research in computer graphics, vision, and machine learning, supported by a novel physics-based reconstruction framework.

Contribution

The paper introduces ScalarFlow, the first large-scale dataset of real smoke flows, and proposes a new physics-based reconstruction method from limited video data.

Findings

The dataset captures complex buoyancy-driven flows with turbulence.

Reconstruction method accurately estimates unseen inflow regions.

High simulation resolution is needed to replicate flow complexity.

Abstract

In this paper, we present ScalarFlow, a first large-scale data set of reconstructions of real-world smoke plumes. We additionally propose a framework for accurate physics-based reconstructions from a small number of video streams. Central components of our algorithm are a novel estimation of unseen inflow regions and an efficient regularization scheme. Our data set includes a large number of complex and natural buoyancy-driven flows. The flows transition to turbulent flows and contain observable scalar transport processes. As such, the ScalarFlow data set is tailored towards computer graphics, vision, and learning applications. The published data set will contain volumetric reconstructions of velocity and density, input image sequences, together with calibration data, code, and instructions how to recreate the commodity hardware capture setup. We further demonstrate one of the many potential application areas: a first perceptual evaluation study, which reveals that the complexity of the captured flows requires a huge simulation resolution for regular solvers in order to recreate at least parts of the natural complexity contained in the captured data.