Abstract
Discovering the 3D atomic structure of molecules sutch as proteins and viruses is a fiundamental research problem in biology and medicine.Electron Cryomicroscopy(Cryo-EM)is a promising vision-based technique for structure es-timation which attempts to reconstruct 3D structures from 2D images.This paper addresses the challenging prob-lem of 3D reconstruction from 2D Cryo-EM images.A new framework for estimation is introduced which relies on modern stochastic optimization techniques to scale to large datasets.We also introduce a novel technique which reduces the cost of evaluating the objective function dur-ing optimization by over five orders or magnitude.The net result is an approach capable of estimating 3D molecular structure from large scale datasets in about a day on a sin-gle workstation.