dtour: a steerable tour de vis through high-dimensional data

TL;DR

dtour is an interactive, GPU-accelerated visualization tool that enables flexible exploration of high-dimensional data through various projection techniques within a browser environment.

Contribution

It introduces a unified interface combining multiple projection exploration methods, scalable to millions of points, and integrates with Python and JavaScript ecosystems.

Findings

Enables exploration of high-dimensional data with millions of points in real-time.

Supports both guided and unrestrained projection navigation.

Validated on text, image, and single-cell datasets for revealing structure and validating reductions.

Abstract

Understanding high-dimensional data requires projecting it into lower-dimensional spaces, but any single projection inevitably loses information or introduces distortions. Tours address this limitation through animation of 2D projection sequences, yet existing tools present tradeoffs in the freedom and steerability of projection traversal, providing little to no ability to move between expert-guided paths and unrestrained exploration. We present dtour, a tour interface that combines static projection previews, reversible scrubbing along continuous geodesic projection paths, manual projection manipulation, and a wandering grand tour, all within a single progressive exploration interface. dtour scales to millions of points via GPU-accelerated rendering, runs in any modern browser, and integrates with both Python and JavaScript ecosystems. We demonstrate dtour on text, image, and single-cell data for two usage scenarios: gradually revealing structure in high-dimensional data and validating non-linear dimensionality reduction outputs.