DeepIcon: A Hierarchical Network for Layer-wise Icon Vectorization
This work addresses a domain-specific challenge in computer graphics for icon vectorization, offering improvements in editing and manipulation utility.
The paper tackles the problem of converting raster images to vector formats, which often results in incomplete shapes and redundant paths, by introducing DeepIcon, a hierarchical network that efficiently generates Scalable Vector Graphics (SVGs) directly from raster images without needing a differentiable rasterizer.
In contrast to the well-established technique of rasterization, vectorization of images poses a significant challenge in the field of computer graphics. Recent learning-based methods for converting raster images to vector formats frequently suffer from incomplete shapes, redundant path prediction, and a lack of accuracy in preserving the semantics of the original content. These shortcomings severely hinder the utility of these methods for further editing and manipulation of images. To address these challenges, we present DeepIcon, a novel hierarchical image vectorization network specifically tailored for generating variable-length icon vector graphics based on the raster image input. Our experimental results indicate that DeepIcon can efficiently produce Scalable Vector Graphics (SVGs) directly from raster images, bypassing the need for a differentiable rasterizer while also demonstrating a profound understanding of the image contents.