Cloud2Curve: Generation and Vectorization of Parametric Sketches

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 avariable-degree Bezier 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 Bezier equivalent. ´We evaluate the generation and vectorization capabilities of our model on Quick, Draw! and K-MNIST datasets.The analysis of free-hand sketches using deep learning[40] has flourished over the past few years, with sketchesnow being well analysed from classification [43, 42] andretrieval [27, 12, 4] perspectives. Sketches for digital analysis have always been acquired in two primary modalities- raster (pixel grids) and vector (line segments). Rastersketches have mostly been the modality of choice for sketchrecognition and retrieval [43, 27]. However, generativesketch models began to advance rapidly [16] after focusing on vector representations and generating sketches assequences [7, 37] of waypoints/line segments, similarly tohow humans sketch. As a happy byproduct, this paradigmleads to clean and blur-free image generation as opposedto direct raster-graphic generations [30]. Recent workshave studied creativity in sketch generation [16], learning tosketch raster photo input images [36], learning efficien