327City and Guilds BuildingSouth Kensington Campus
//
Summary
Publications
Citation
BibTex format
@inproceedings{Kelshaw:2022, author = {Kelshaw, D and Rigas, G and Magri, L}, title = {Physics-informed CNNs for super-resolution of sparse observations on dynamical systems}, url = {http://arxiv.org/abs/2210.17319v2}, year = {2022} }
TY - CPAPER AB - In the absence of high-resolution samples, super-resolution of sparseobservations on dynamical systems is a challenging problem with wide-reachingapplications in experimental settings. We showcase the application ofphysics-informed convolutional neural networks for super-resolution of sparseobservations on grids. Results are shown for the chaotic-turbulent Kolmogorovflow, demonstrating the potential of this method for resolving finer scales ofturbulence when compared with classic interpolation methods, and thuseffectively reconstructing missing physics. AU - Kelshaw,D AU - Rigas,G AU - Magri,L PY - 2022/// TI - Physics-informed CNNs for super-resolution of sparse observations on dynamical systems UR - http://arxiv.org/abs/2210.17319v2 UR - http://hdl.handle.net/10044/1/101242 ER -