Imaging Lunch: Understanding Deep Neural Networks for Imaging Applications

Our in-house expert Daniel Sage will give a workshop on Understanding Deep Neural Networks for Imaging Applications during our next imaging lunch. Open to all EPFL PhD students and Postdocs!

Artificial neural networks, particularly deep convolutional neural network (CNN) architectures, have become critical and ubiquitous in image analysis. CNN are particularly well-suited for handling complex image data. Among these, the U-Net—the most iconic CNN architectures—has proven highly effective for tasks such as deblurring, restoration, detection, and segmentation.

To demystify CNN, we will revisit the foundational image-processing principles on which they are built, including convolution, non-linear filtering, cascading filters, shift-invariance, multiresolution, and local feature extraction. Finally, we will showcase CNNs in imaging applications and highlight key design considerations, including the importance of suitable training data.

Audience: Scientists, engineers, and general public with an interest in imaging. No programming or mathematical knowledge is required.

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!