Model Based Deep Learning with Application to Super Resolution - IPAM at UCLA

Explore a 47-minute lecture on model-based deep learning and its application to super-resolution imaging. Delve into the integration of parametric models with optimization tools and classical algorithms, leading to efficient and interpretable networks that require smaller training sets. Discover how these concepts are applied to super-resolution microscopy, enabling high-quality imaging from limited high-emitter density frames without prior knowledge of the optical system. Gain insights into overcoming the black-box nature of deep neural networks while maintaining their performance advantages in signal and image processing.