Enhancing Frequency Response in Implicit Full Waveform Inversion via Fourier Encoding (opens in new tab)

Implicit full waveform inversion (IFWI) has emerged as a promising framework for reconstructing complex subsurface structures. However, conventional methods suffer from a fundamental bias conflict, in which the inherent spectral bias of neural networks, such as the low-frequency bias of multilayer perceptrons (MLPs) and the high-frequency bias of sinusoidal representation networks (SIRENs), conflicts with the physics-based gradient’s requirement for progressive frequency reconstruction. To re...