Topology ToolKit

Events

Topological Analysis of Ensemble Scalar Data with TTK

IEEE VIS 2022 Tutorial
Monday, October 17, 9:00am-12:25am (CT)
IEEE VIS, Oklahoma City, OK, USA

Registration:
 · Tutorial pre-registration
 · IEEE VIS 2022 registration


Please see our Events page for all tutorials in this series.

Note to attendees

 ·  We kindly ask potential attendees to pre-register at https://forms.gle/9b7TTERsjMs49g9m8 in order for us to reach out to them ahead of the tutorial with information updates (for instance, last minute updates, instructions for the download of the tutorial material package, etc.)
 ·  If you plan to attend this tutorial, we invite you to install TTK on your system before the tutorial. Several installation alternatives are available (via ParaView's official binaries, sources, Ubuntu packages, Anaconda packages, Docker images, VirtualBox images, etc.). See the installation instructions.
 ·  Please do not hesitate to reach out ahead of the tutorial if you experience any issue installing TTK: ttk-users@googlegroups.com

Resources and Materials

 ·  Installation instructions
 ·  Tutorial data package
 ·  Detailed tutorial submission document

Abstract

This tutorial presents topological methods for data analysis and visualization from a user's perspective, with the Topology ToolKit (TTK), an open-source library for topological data analysis. In particular, similarly to 2021, this year's tutorial has a special focus on ensemble data analysis with TTK, but with an updated content. Topological methods have gained in popularity and maturity over the last twenty years and success stories of established methods have been documented in a wide range of applications (combustion, chemistry, astrophysics, material sciences, etc.) with both acquired and simulated data, in both post-hoc and in-situ contexts. This tutorial provides a beginner's introduction to topological methods for practitioners, researchers, students, and lecturers, with a special emphasis towards ensemble data analysis. In particular, instead of focusing on theoretical aspects and algorithmic details, this tutorial focuses on how topological methods can be used in practice to reduce ensemble datasets into concise yet meaningful topological data representations and how these representations can support advanced analysis. The tutorial describes in detail how to achieve these tasks with TTK. In contrast to the first iterations of this tutorial, this iteration focuses on the specific usage of TTK for ensemble data analysis, similarly to the 2021 edition, but with an updated content, including updated or additional features for ensemble data processing. First, we provide a general introduction to topological methods and their application in data analysis, and a brief overview of TTK's main entry point for end users, namely ParaView, will be presented. Second, we detail TTK's software infrastructure for ensemble data analysis, including TTK's Docker support (to facilitate its deployment on computing servers), a tour of the topological data representations supported by TTK, and lastly TTK's cinema support (to manipulate ensemble of topological data representations with a database formalism). Third, we will present concrete use cases of ensemble data analysis and visualization, using contour tree alignment as well as ensemble clustering and summarization with persistence diagrams and merge trees. Presenters of this tutorial include experts in topological methods, core authors of TTK as well as active users, coming from academia and industry. This tutorial mostly targets students, practitioners and researchers who are not necessarily experts in topological methods but who are interested in using them in their daily tasks. We also target researchers already familiar to topological methods and who are interested in using or contributing to TTK. We kindly ask potential attendees to optionally pre-register at the following address, in order for us to reach out to them ahead of the tutorial with information updates (for instance, last minute updates, instructions for the download of the tutorial material package, etc.): https://forms.gle/9b7TTERsjMs49g9m8 .

Organizers

Schedule

  1. 9:00am Preliminaries (70 minutes)

  2. 10:10am Beak, discussion and Q&A (15 minutes)

  3. 10:25am TTK infrastructure for ensemble data analysis (60 minutes)

  4. 11:25am Beak, discussion and Q&A (15 minutes)

  5. 11:40am Ensemble processing with TTK (50 minutes)

Talks

Preliminaries

 ·  General introduction (Julien Tierny)
 ·  Introduction to topological methods for data analysis (Bei Wang)
This talk will present the core tools in topological data analysis (critical points, Persistence diagrams, Reeb graphs and their variants, annd Morse-Smale complexes). In particular, it will detail how these tools can be used for data segmentation and feature extraction.
 ·  Quick introduction to ParaView's user interface (Charles Gueunet)
This talk will provide a brief description of ParaView's main interface, in order to support its usage for beginners in the subsequent hands-on session. This will cover the usage of filters, pipeline design and view manipulation, state files backups, Python exports, and a quick introduction to Catalyst (for in-situ computations).

TTK infrastructure for ensemble data analysis

 ·  Running TTK with Docker (Christoph Garth)
This talk will describe how to use TTK's most recent Docker support, which is useful for deploying TTK on high-performance infrastructures (for the in-situ computation of topological signatures).
 ·  A tour of TTK (Joshua Levine)
This talk will provide an overview of the topological signatures supported by TTK and will illustrate how to apply them on concrete examples.
 ·  TTK Compact Triangulation (Federico Iuricich)
This talk will introduce TTK's novel compact triangulation data structure, specialized for the handling of large datasets.
 ·  TTK Cinema support (Jonas Lukasczyk)
This talk will introduce TTK’s Cinema support for storing and manipulating databases of topological signatures representing an ensemble.

Ensemble processing with TTK

 ·  Contour tree alignment (Florian Wetzels)
This talk will introduce the contour tree alignment module of TTK, which enables the simultaneous and coherent planar layout of an ensemble of contour trees (representing an ensemble dataset) for visual inspection purposes.
 ·  Distances, barycenters and clusters of Persistence diagrams (Jules Vidal)
This talk will show how to efficiently compare and cluster the members of an ensemble based on their persistence diagram.
 ·  Distances, barycenters and clusters of merge trees (Mathieu Pont)
This talk will show how to efficiently compare and cluster the members of an ensemble based on their merge tree.
 ·  Ensemble summarization with linked planar views (Pierre Guillou)
This talk will show step-by-step how to generate, with multidimensional scaling, planar summarizations of the input ensemble, where each member is represented as a point in 2D and where similar members (in terms of their topological signatures) are projected in similar locations. Linked views, related to each point of the summarization, enable the visual inspection of each topological signature. Together, the planar summarization and its detailed linked view enable the visual analysis of global trends in the ensemble.
 ·  Concluding remarks (Julien Tierny)

Pictures of the tutorial

Additional Material

 ·  TTK overview video
 ·  TTK example website (scripts, state files, screenshots, etc.)
 ·  TTK Technical presentation at IEEE VIS 2017 (for advanced users)
 ·  TTK Technical report (for advanced users)
 ·  TTK feature overview
 ·  TTK user forum
 · Using TTK in-situ with Catalyst: Online tutorial
 · Using TTK with Python: Online example - Online tutorial
 · Using TTK with VTK/C++: Online example
 · Using TTK with pure C++: Online example
 · Extending TTK with a new module: Online tutorial (computation of Betti numbers)
 · TTK developer documentation