![]() ![]() Point set registration: Coherent point drift. HelpingHand: Example-based stroke stylization. DecoBrush: Drawing structured decorative patterns by example. RealBrush: Painting with examples of physical media. Fully convolutional networks for semantic segmentation. Style-preserving English handwriting synthesis. Font generation of personal handwritten Chinese characters. Automatic generation of large-scale handwriting fonts via style learning. Automatic shape morphing for Chinese characters. Computationally evaluating and synthesizing Chinese calligraphy. Journal of Machine Learning Research 6, Nov (2005), 1783-1816. Probabilistic non-linear principal component analysis with Gaussian process latent variable models. Human-level concept learning through probabilistic program induction. Computer Processing of Oriental Languages 10, 3 (1996), 307-323. A heuristic search approach to Chinese glyph generation using hierarchical character composition. In Advances in Neural Information Processing Systems 25. ImageNet classification with deep convolutional neural networks. Farsi font recognition based on Sobel-Roberts features. Study of several handwritten Chinese character directional feature extraction approaches. IEEE Transactions on Pattern Analysis and Machine Intelligence 12, 6 (1990), 541-551. Analysis of thinning algorithms using mathematical morphology. Image-to-image translation with conditional adversarial nets. Reducing the dimensionality of data with neural networks. Deep residual learning for image recognition. ![]() ACM Transactions on Graphics (TOG) 35, 3 (2016), 26. Controlling perceptual factors in neural style transfer. A method of computerizing the calligraphical rules basing on CC structure code. Intelligent Chinese character design and an experimental system ICCDS. Statistic model-based simulation on calligraphy creation. Analysis and modeling of naturalness in handwritten characters. An automatic stroke extraction method using manifold learning. ACM Transactions on Graphics 33, 4 (2014), 91. Analyzing 50k fonts using deep neural networks. ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games. ![]() It can also be observed from our experiments that recently-popularized deep learning based end-to-end methods are not able to properly handle this task, which implies the necessity of expert knowledge and handcrafted rules for many applications. Using our system, for the first time, the practical handwriting font library in a user’s personal style with arbitrarily large numbers of Chinese characters can be generated automatically. Experiments including Turing tests with 97 participants demonstrate that the proposed system generates high-quality synthesis results, which are indistinguishable from original handwritings. Second, we develop a set of novel techniques to learn and recover users’ overall handwriting styles and detailed handwriting behaviors. First, we design an effective stroke extraction algorithm that constructs best-suited reference data from a trained font skeleton manifold and then establishes correspondence between target and reference characters via a non-rigid point set registration approach. Major technical contributions of our system are twofold. To solve this problem, we propose a system, EasyFont, to automatically synthesize personal handwriting for all (e.g., Chinese) characters in the font library by learning style from a small number (as few as 1%) of carefully-selected samples written by an ordinary person. Consistently and correctly writing out such huge amounts of characters is usually an impossible mission for ordinary people. For example, the official standard GB18030-2000 for commercial font products consists of 27,533 Chinese characters. Generating personal handwriting fonts with large amounts of characters is a boring and time-consuming task.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |