Seismic compressive sensing by generative inpainting network: Toward an optimized acquisition survey

Abstract The use of deep learning models as priors for compressive sensing tasks presents new potential for inexpensive seismic data acquisition. Conventional recovery usually suffers from undesire...