Projects

Proteus is a multi-agent framework designed to automatically adapt desktop visualizations for mobile devices.

We proposed a visual design creativity activation model.

We proposed a spatial constraint model for activating static visualization.

We proposed an adaptive data management architecture for the real-time visualization of spatiotemporal datasets

A system that supports interactive data communication by leveraging AI to help users generate, refine, and present data-driven narratives.

We proposed a deep learning-based volume rendering pipeline.

We designed and trained deep learning models to generate short descriptions for SVG-formatted charts automatically.

An interactive system that helps users explore and refine prompts for text-to-image generation through visual and iterative workflows.

We proposed a semi-automatic method for title generation.

OldVisOnline is a collection of old visualizations published before 1950. We have been studying a variety of usage scenarios of OldVisOnline.

We designed a visual analytics system for analyzing the impact of different asteroids.

We designed a deep learning-based method for volume rendering best view selection.