| CONTENTS |
|
Seminars |
Learning continuous and piecewise-linear functions and measuring their complexity29 Mar 2021
Continuous-Domain Formulation of Inverse Problems for Composite Sparse Plus Smooth Signals08 Mar 2021
Graphic: Graph-Based Hierarchical Clustering for Single-Molecule Localization Microscopy01 Mar 2021
Inverse problems for image-based characterisation of cellular mechanics: how do cells move?23 Feb 2021
Optical Diffraction Tomography with Single-Molecule Localization Microscopy08 Feb 2021
Recent algorithmic advances in Phase Retrieval22 Dec 2020
Robust and Sparse Regression Models for One-Dimensional Data08 Dec 2020
Shortest Multi-Spline Bases for Generalized Sampling24 Nov 2020
A Hybrid Stochastic Framework for Signal Recovery10 Nov 2020
Wavelets in harmonic analysis and signal processing27 Oct 2020
Optimal transport-based metric for single-molecule localization microscopy (SMLM)20 Oct 2020
Time-dependent deep image prior for dynamic MRI08 Sep 2020
Shortest Multi-spline Bases for Generalized Sampling03 Aug 2020
Convex Optimization in Infinite Sums of Banach Spaces Using Besov Regularization13 Jul 2020
Measuring Complexity of Deep Neural Networks29 Jun 2020
Robust Phase Unwrapping via Deep Image Prior for Quantitative Phase Imaging22 Jun 2020
Space Varying Blurs: Estimation, Identification and Applications18 May 2020
Matrix factorization and phase retrieval for deep fluorescence microscopy11 May 2020
CryoGAN: A New Reconstruction Paradigm for Single-particle Cryo-EM Via Deep Adversarial Learning27 Apr 2020
In this talk, we present CryoGAN, a new paradigm for single-particle cryo-EM reconstruction based on unsupervised deep adversarial learning. The major challenge in single-particle cryo-EM is that the measured particles have unknown poses. Current reconstruction techniques either estimate the poses or marginalize them away—steps that are computationally challenging. CryoGAN sidesteps this problem by using a generative adversarial network (GAN) to learn the 3D structure whose simulated projections most closely match the real data in a distributional sense. The architecture of CryoGAN resembles that of standard GAN, with the twist that the generator network is replaced by a cryo-EM physics simulator. CryoGAN is an unsupervised algorithm that only demands picked particle images and CTF estimation as inputs; no initial volume estimate or prior training are needed. Moreover, it requires minimal user interaction and can provide reconstructions in a matter of hours on a high-end GPU. The current results on synthetic datasets show that the CryoGAN can reconstruct a high-resolution volume with its adversarial learning scheme. Preliminary results on real β-galactosidase data demonstrate its ability to capture and exploit real data statistics in more challenging imaging conditions. If the time permits, we would also like to discuss its extension for multiple conformations.
CryoGAN: A New Reconstruction Paradigm for Single-particle Cryo-EM Via Deep Adversarial Learning27 Apr 2020
Gibbs Sampling-Based Statistical Inference for Inverse Problems20 Apr 2020
Robust Reconstruction of Fluorescence Molecular Tomography With An Optimized Illumination Pattern04 Mar 2020
Solving various domain translation problems using deep convolutional framelets11 Feb 2020
Adaptive regularization for three-dimensional optical diffraction tomography17 Dec 2019
About the use of non-imaging data to improve domain adaptation for spinal cord segmentation on MRI26 Nov 2019
Lagrangian Tracking of Bubbles Entrained by a Plunging Jet19 Nov 2019
Multigrid Methods for Helmholtz equation and its application in Optical Diffraction Tomography05 Nov 2019
Efficient methods for solving large scale inverse problems17 Oct 2019
© 2010 EPFL • webmaster.big@epfl.ch • 26.01.2010