Segmentation-guided Layer-wise Image Vectorization with Gradient Fills
This addresses the problem of creating high-quality vector graphics with gradients for users in graphic design and digital art, though it is incremental as it builds on existing vectorization techniques.
The paper tackles the challenge of filling vector primitives with gradients in raster-to-vector conversion by proposing a segmentation-guided framework that uses gradient-aware segmentation and Bézier paths to produce vector graphics with radial gradient fills. The result is improved visual quality and layer-wise topology compared to prior methods, as demonstrated on various inputs without reliance on datasets.
The widespread use of vector graphics creates a significant demand for vectorization methods. While recent learning-based techniques have shown their capability to create vector images of clear topology, filling these primitives with gradients remains a challenge. In this paper, we propose a segmentation-guided vectorization framework to convert raster images into concise vector graphics with radial gradient fills. With the guidance of an embedded gradient-aware segmentation subroutine, our approach progressively appends gradient-filled Bézier paths to the output, where primitive parameters are initiated with our newly designed initialization technique and are optimized to minimize our novel loss function. We build our method on a differentiable renderer with traditional segmentation algorithms to develop it as a model-free tool for raster-to-vector conversion. It is tested on various inputs to demonstrate its feasibility, independent of datasets, to synthesize vector graphics with improved visual quality and layer-wise topology compared to prior work.