Volumetric reconstruction of drop impact dynamics by means of color-coded glare points and deep learning

The present work introduces a method for the volumetric reconstruction of the gas-liquid interface of a droplet impinging on a solid surface from experimental image data by means of deep learning.
Differently colored glare points are used in order to encode three-dimensional information of the phase boundary on the image in addition to a standard shadowgraph.
It is demonstrated that this volumetric information can be successfully reconstructed into a three-dimensional representation of the droplet phase boundary by the proposed neural network.