On the Interface of Optimization and Deep Learning for Computational Imaging
Abstract:
This session focuses on computational imaging methods at the interface of optimization and deep learning. Recent advances aiming at pairing these two areas has led to significant advancements in image reconstruction and enhancement. Optimization techniques traditionally provided mathematically rigorous solutions for imaging problems, but they often struggled with high-dimensional data and non-linearities. Deep learning, particularly neural networks, has transformed this landscape by learning complex mappings from data, enabling faster and higher expressivity in image solutions. When combined, these approaches harness optimization algorithms to refine and guide the training of deep networks, resulting in robust models that are highly efficient for performing computational imaging tasks.
Organiser & chair: Audrey Repetti
Session Schedule
- 17:30 - 17:50
- Invited talk: On the Interface of Optimization and Deep Learning for Computational Imaging
- Audrey Repetti
- 17:50 - 18:10
- Invited talk: True and False Monotone Neural Networks
- Jean-Christophe Pesquet
- 18:10 - 18:30
- Invited talk: Data-Driven Spatial Adaptivity for Regularising Inverse Problems
- Sebastian Neumayer
- 18:30 - 18:50
- Invited talk: Deep Image Regularisation for Poisson Inverse Problems via Mirror Descent
- Luca Calatroni
- 18:50 - 19:10
- Invited talk: (Deep) Machine Learning for Exoplanet Detection in Direct Imaging at High Contrast
- Julien Mairal
- 19:10 - 20:30
- Invited poster: Convergence Analysis of Plug-and-Play Methods for Image Inverse Problems
- Samuel Hurault*
- 19:10 - 20:30
- Invited poster: Stochastic Deep Restoration Priors for Imaging Inverse Problems
- Chicago Park
- 19:10 - 20:30
- Contributed poster: LoFi: Scalable Local Image Reconstruction with Implicit Neural Representation
- Amirehsan Khorashadizadeh*
- 19:10 - 20:30
- Contributed poster: Equivariant Plug-and-play Image Reconstruction
- Thomas Moreau
- 19:10 - 20:30
- Contributed poster: Variational Reconstruction of Parameter Maps in Medical Imaging and Materials Science
- Gabriele Scrivanti*