Supporting Expressive and Faithful Pictorial Visualization Design with Visual Style Transfer

Yang Shi, Pei Liu, Siji Chen, Mengdi Sun, Nan Cao

View presentation: 2022-10-19T15:57:00Z GMT-0600 Change your timezone on the schedule page
2022-10-19T15:57:00Z
Exemplar figure, described by caption below
Vistylist is a design support tool that disentangles the visual style of a source pictorial visualization from its content and transfers the visual style to one or more intended pictorial visualizations. The user interface of Vistylist consists of four main components: (1) users upload their source visualizations to the Design Panel (A) to extract visual elements, (2) users then upload their datasets to the Data Panel (B) to create field-channel mappings, (3) users can view the results on the Canvas (C), where they can adjust icons using the Icon Panel (D), and (4) the Suggestion Panel (E) suggested alternative designs.

Prerecorded Talk

The live footage of the talk, including the Q&A, can be viewed on the session page, Visualization Opportunities.

Fast forward
Abstract

Pictorial visualizations portray data with figurative messages and approximate the audience to the visualization. Previous research on pictorial visualizations has developed authoring tools or generation systems, but their methods are restricted to specific visualization types and templates. Instead, we propose to augment pictorial visualization authoring with visual style transfer, enabling a more extensible approach to visualization design. To explore this, our work presents Vistylist, a design support tool that disentangles the visual style of a source pictorial visualization from its content and transfers the visual style to one or more intended pictorial visualizations. We evaluated Vistylist through a survey of example pictorial visualizations, a controlled user study, and a series of expert interviews. The results of our evaluation indicated that Vistylist is useful for creating expressive and faithful pictorial visualizations.