Jiang et al. (2021) | Recognizing Vector Graphics without Rasterization¶
Jiang et al. do not work on generative models for vector graphics. But they focus on the fundamental recognition tasks of object localization and classification within vector graphics – without rendering the graphic to pixels.
This paper is especially interesting due to the chosen data representation as a graph and the use of Graph Neural Networks (GNN).
Available resources at a glance
Fig. 63 Screenshot of the Jiang et al. paper – the overall pipeline of the proposed method¶