Multimodal Image Analysis Workshop

Multimodal Image Analysis Workshop

Come join us for a hands-on workshop on multimodal image analysis at the QUT Gardens Point Campus organised by Planetary Surface Exploration group! Dive into the world of image processing, computer vision, and machine learning in this interactive event. Learn from experts in the field and get valuable insights into the latest techniques and tools. Whether you're a beginner or an experienced researcher, this workshop is perfect for anyone interested in advancing their skills in image analysis. Don't miss out on this exciting opportunity to expand your knowledge and network with like-minded individuals!

Below are some expected outcomes:

Deeper Understanding of Multimodal Image Analysis - Attendees will leave the workshop with a deeper understanding of the latest techniques and tools in multimodal image analysis.

Cross-Disciplinary Learning and Adaptation - Participants will explore how techniques can be adapted or integrated into their own work, regardless of scale or original application area.

Identification of Key Challenges and Innovative Approaches - Participants will identify key challenges in multimodal image analysis (specific to thier fields), and brainstorm innovative approaches or solutions. This could involve practical issues such as co-registration, segmentation, or dimensionality reduction of multimodal image stacks.

Fostering Collaborations and Knowledge Exchange - The workshop will foster connections among attendees that could lead to future collaborations, joint research projects, or ongoing knowledge exchange, promoting a network for ongoing dialogue and support in multimodal image analysis.

Actionable Insights and Strategies - Participants will leave with actionable insights or strategies for integrating multimodal image analysis techniques into their work. This includes practical takeaways, such as tools and approaches (e.g., using QuPath, performing dimensionality reduction) that can be immediately applied in their research.

Workshop will include two parts:

Part 1: Presentation Block on Multimodal Registration and Segmentation (45-60 mins)Goal - Provide a real-world grounding in the challenges attendees face. 5-minute presentations where participants explain their specific needs for registering multimodal imagery (typically from different instruments) and performing semantic segmentation on combined imagery.

Hands on demo: 30-minute session using QuPath to demonstrate practical workflows for image segmentation and analysis. Attendees will learn how to move beyond using an optical image merely as context to read a spectral or chemical image and integrate various modalities meaningfully.

Part 2: Dimensionality Reduction

Goal - Rather than stacking multiple co-registered images that can overwhelm the human brain, explore ways to summarize these layers into far fewer, more manageable layers (e.g., RGB images where each color channel contains specific information).

5-minute presentations of participants' challenges with x-dimensional stacks and their needs for dimensionality reduction — do we ask people to come prepared with their slides or have it impromptu 5 min presentation?

A brainstorming session to allow for group collaboration to develop innovative solutions.

Pre-workshop Preparation

We will share a short video to registered participants prior to the workshop as a pre-workshop preparation.

Location:

P Block, Room 413

QUT Gardens Point

Brisbane, QLD, 4001

*Banner image: A digital microscope photomicrograph of a rock from the deep mantle (125 km). Earth scientists use the vivid colours in polarised light to identify minerals | Image credit: Balz Kamber