Machine Learning for 3D Optical Imaging - IPAM at UCLA
Institute for Pure & Applied Mathematics (IPAM) via YouTube
Overview
Explore machine learning applications in 3D optical imaging through this 38-minute lecture by Demetri Psaltis from École Polytechnique Fédérale de Lausanne (EPFL). Delve into optical diffraction tomography (ODT) techniques for retrieving 3D object shapes from multiple holographic 2D projections. Discover how machine learning enhances ODT performance by correcting reconstruction distortions caused by limited projections. Examine two approaches: one utilizing digital phantoms as ground truth for neural network training, and another leveraging physics-based constraints to produce improved 3D images. Recorded at IPAM's Diffractive Imaging with Phase Retrieval Workshop at UCLA, this talk offers valuable insights into cutting-edge optical imaging techniques.
Syllabus
Demetri Psaltis - Machine Learning for 3D Optical Imaging - IPAM at UCLA
Taught by
Institute for Pure & Applied Mathematics (IPAM)