The Para2Mesh framework of iterative denoising to accomplish flow field reconstruction and adaptive mesh moving (IMAGE)
Caption
Para2Mesh utilizes an end-to-end dual-diffusion framework consisting of a flow field diffusion network and a mesh diffusion network, the conditions of both of which include design parameters and mesh structure information. The diffusion process in both networks gradually adds noise to the clean samples and eventually transforms them into Gaussian noise. In the denoising process, the random Gaussian noise is gradually denoised until a reconstructed flow field and an adaptive mesh consistent with the flow structure are generated.
Credit
Chinese Journal of Aeronautics
Usage Restrictions
News organizations may use or redistribute this image, with proper attribution, as part of news coverage of this paper only.
License
Original content