Imaging Lunch: Extending Neural Networks Beyond 2D: Approaches for 3D and Temporal Data
While most neural network-based image analysis focuses on 2D data, many scientific fields rely on multidimensional data, such as 3D volumes from microscopy or tomography. This lunch session, led by our image analysis expert Florian Aymanns, delves into the challenges and solutions for applying neural networks to 3D and temporal datasets.
The session will begin with a discussion of the unique challenges of processing 3D data using neural networks, followed by an overview of existing methods tailored for such data.
Participants will explore techniques to adapt 2D neural network approaches for 3D datasets and strategies for handling temporal data effectively.
The event also includes a collaborative discussion, where participants can share their experiences and insights, fostering peer learning and exchange.
Audience:
This session is ideal for scientists who work with 3D or higher-dimensional data and are interested in leveraging neural networks for analysis.
Learning Objectives:
- Understand the challenges of processing 3D data with neural networks.
- Gain an overview of methods for analyzing 3D and temporal data, including hybrid approaches using 2D techniques.
Prerequisites:
- Basic understanding of neural networks.
- Experience or interest in analyzing 3D or higher-dimensional data.
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!
