MARVEL: Raster Manga Vectorization via Primitive-wise Deep Reinforcement Learning
Authors:
- Hao SuHSHao Su
- Xuefeng LiuXLXuefeng Liu
- Jianwei NiuJianwei Niu
- Jiahe CuiJCJiahe Cui
- Ji WanJWJi Wan
- Xinghao WuXWXinghao Wu
- Nana WangNWNana Wang
Preprint
Abstract
Manga is a fashionable Japanese-style comic form that is composed of black-and-white strokes and is generally displayed as raster images on digital devices. Typical mangas have simple textures, wide lines, and few color gradients, which are vectorizable natures to enjoy the merits of vector graphics, e.g., adaptive resolutions and small file sizes. In this paper, we propose MARVEL (MAnga's Raster to VEctor Learning), a primitive-wise approach for vectorizing raster mangas by Deep Reinforcement Learning (DRL). Unlike previous learning-based methods which predict vector parameters for an entire image, MARVEL introduces a new perspective that regards an entire manga as a collection of basic primitives\textemdash stroke lines, and designs a DRL model to decompose the target image into a primitive sequence for achieving accurate vectorization. To improve vectorization accuracies and decrease file sizes, we further propose a stroke accuracy reward to predict accurate stroke lines, and a pruning mechanism to avoid generating erroneous and repeated strokes. Extensive subjective and objective experiments show that our MARVEL can generate impressive results and reaches the state-of-the-art level. Our code is open-source at: https://github.com/SwordHolderSH/Mang2Vec.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Deep Line Drawing Vectorization via Line Subdivision and Topology ReconstructionComputer Graphics Forum, 2019
- Deep extraction of manga structural linesACM Transactions on Graphics, 2017
- Manga Vectorization and Manipulation with Procedural Simple ScreentoneIEEE Transactions on Visualization and Computer Graphics, 2016
- Closure-aware sketch simplificationACM Transactions on Graphics, 2015
- Topology-driven vectorization of clean line drawingsACM Transactions on Graphics, 2013
- Smart Scribbles for Sketch SegmentationComputer Graphics Forum, 2012
- Vectorizing Cartoon AnimationsIEEE Transactions on Visualization and Computer Graphics, 2009
- Richness-preserving manga screeningACM Transactions on Graphics, 2008
- Deringing cartoons by image analogiesACM Transactions on Graphics, 2006
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004