Viva: a Variational Image Vectorization Algorithm on Dual-Primal Graph Pairs

We propose a novel variational image vectorization algorithm (VIVA) which alternatively smooths contours by affine shortening flow and eliminates spurious regions by minimizing a Mumford-Shah-type functional. We introduce dual-primal graphs representing domain partitions which allows for effective iterative computation. The method provides varying levels of simplicity on the topology of the resulted vector graphics while effectively removing pixelization. It compares favorably to the state-of-the-art (SOTA) vectorization methods.

1. INTRODUCTION

Image vectorization is the process of converting raster images to vector graphics [1], [2]. Upon geometric transformations, raster images lose contrast and resolution due to grid resampling, whereas vector graphics keep resolution and contrast intact. Vectorizing images thus fosters wide applications in areas such as graphic design [3] and font editing [4].