VisImages: a Corpus of Visualizations in the Images of Visualization Publications

Images in visualization publications contain rich information, e.g., novel visualization designs and common combinations of visualizations. A systematic collection of these images can contribute to the community in many aspects, such as literature analysis and automated tasks for visualization. In this paper, we build and make public a dataset, VisImages, which collects 12,267 images with captions from 1,397 papers in IEEE InfoVis and VAST. Based on a refined taxonomy for visualizations in publications, the dataset includes 35,096 annotated visualizations, as well as their positions. We demonstrate the usefulness of VisImages through three use cases: 1) exploring and analyzing the evolution of visualizations with VisImages Explorer, 2) training and benchmarking models for visualization classification, and 3) localizing and recognizing visualizations in the images automatically.