Projects
I’m a PhD candidate specializing in developing computational approaches to manipulate emotion, semantics, and aesthetics in data-driven graphics. My background also includes work in natural language interfaces for visualizations and dynamic network visualization from my undergraduate studies.
Intuitively understand LLM-generated code with animated visualizations and scrollytelling. Interact with the interactive diagram to steer the data analysis process.
Bringing interactive visualization to life with VLM-powered motion design.
Create kinetic typography with motion transfer. By deforming text while maintaining its legibility, semantics/emotions can be expressed through animation.
An approach to generate word clouds that conveys nuanced emotions. The animation scheme of kinetic typography is propagated to multiple words.
Encode hierarchical text as a dual resolution wordle to bring the context together with keywords. This is the topic of my undergraduate thesis and I’d love to explore more.
GeoCamera is an authoring tool for specifying camera effects in geospatial data videos according to narrative purposes
LQ2 quantifies the aesthetics of chart layout design by collecting paired crowdsourced comparison data and training a siamness network.
AutoCaption generates short descriptions in natural language for SVG-based charts automatically.
MeasureFlow presents a series of graph measures and offers an overview of dynamic networks to facilitate explorations. It is an extension of Vistorian.
PIMA is a new visualization for dynamic networks. It encodes group features onto a sequence of matrices with pixel technique and supports interactions in a top-down manner.