Res-U2Net: Untrained Deep Learning for Phase Retrieval and Image Reconstruction

arXiv:2404.06657v1 Announce Type: cross
Abstract: Conventional deep learning-based image reconstruction methods require a large amount of training data which can be hard to obtain in practice. Untrained deep learning methods overcome this limitation by training a network to invert a physical model of the image formation process. Here we present a novel untrained Res-U2Net model for phase retrieval. We use the extracted phase information to determine changes in an object's surface and generate a mesh representation of its 3D structure. …

abstract arxiv cs.cv data deep learning eess.iv image network physics.optics practice process retrieval training training data type