Synthetically Trained Icon Proposals for Parsing and Summarizing Infographics
This work addresses the challenge of parsing infographics for applications in news, business, and education, enabling multi-modal summarization through icon detection, classification, and text extraction, but it is incremental as it builds on existing data and methods for a specific domain.
The paper tackled the problem of automatically identifying stand-alone visual elements (icons) in infographics, where existing computer vision methods trained on natural images perform poorly, by proposing a synthetic data generation strategy using Internet-scraped icons to train an icon proposal mechanism, achieving 38% precision and 34% recall on a test set of 1K annotated infographics, compared to 14% precision and 7% recall for the best natural image-trained model.
Widely used in news, business, and educational media, infographics are handcrafted to effectively communicate messages about complex and often abstract topics including `ways to conserve the environment' and `understanding the financial crisis'. Composed of stylistically and semantically diverse visual and textual elements, infographics pose new challenges for computer vision. While automatic text extraction works well on infographics, computer vision approaches trained on natural images fail to identify the stand-alone visual elements in infographics, or `icons'. To bridge this representation gap, we propose a synthetic data generation strategy: we augment background patches in infographics from our Visually29K dataset with Internet-scraped icons which we use as training data for an icon proposal mechanism. On a test set of 1K annotated infographics, icons are located with 38% precision and 34% recall (the best model trained with natural images achieves 14% precision and 7% recall). Combining our icon proposals with icon classification and text extraction, we present a multi-modal summarization application. Our application takes an infographic as input and automatically produces text tags and visual hashtags that are textually and visually representative of the infographic's topics respectively.