Cloud2Curve: Generation and Vectorization of Parametric Sketches
This work addresses the need for efficient and scalable sketch representation in deep learning applications, offering a novel approach for artists and designers, though it builds incrementally on existing inverse graphics and generative modeling techniques.
The paper tackles the problem of generating and vectorizing sketches by modeling them as sequences of parametric Bézier curves instead of raster graphics or waypoint sequences, achieving scalable high-resolution vector sketches that can be trained end-to-end using point-cloud data alone.
Analysis of human sketches in deep learning has advanced immensely through the use of waypoint-sequences rather than raster-graphic representations. We further aim to model sketches as a sequence of low-dimensional parametric curves. To this end, we propose an inverse graphics framework capable of approximating a raster or waypoint based stroke encoded as a point-cloud with a variable-degree Bézier curve. Building on this module, we present Cloud2Curve, a generative model for scalable high-resolution vector sketches that can be trained end-to-end using point-cloud data alone. As a consequence, our model is also capable of deterministic vectorization which can map novel raster or waypoint based sketches to their corresponding high-resolution scalable Bézier equivalent. We evaluate the generation and vectorization capabilities of our model on Quick, Draw! and K-MNIST datasets.