Imaging Lunch: DeepInverse: a PyTorch library for solving imaging inverse problems with deep learning

Thumbnail

Event details

Date 10.12.2025
Hour 11:0013:00
Speaker Dr Julian Tachella, CNRS & ENS de Lyon
Location
Category Conferences - Seminars
Event Language English
Abstract 
In this tutorial, we will introduce the main functionalities of the deepinverse open-source library (https://deepinv.github.io/) through a live demo. The library covers all crucial steps in image reconstruction from the efficient implementation of forward operators (e.g., optics, MRI, tomography), to the definition and resolution of variational problems and the design and training of advanced neural network architectures. You will be able to choose one mini-project to try out the library, such as defining a new forward operator, building a reconstruction method or training a reconstruction network.

Learning Objectives 
  • Understand how to solve imaging inverse problems with state-of-the-art learning-based methods using the library (diffusion models, plug-and-play, self-supervised learning, etc.)
  • Grasp the main functionalities of the library.

Prerequisites 
  • Bring your laptop and install the library by following the instructions here in your preferred configuration (local, ssh server, etc.).
  • Have some experience with imaging inverse problems
Level: advanced

Registration required

About the Imaging Lunches: Once per month, the EPFL Center for Imaging organises an event dedicated to all PhD students and postdocs working with/in imaging. Discuss the latest advances in imaging. Connect with imaging peers. Learn about popular imaging tools!

Links

Practical information

  • General public
  • Free

Organizer

  • EPFL Center for Imaging  

Contact

Share