H-Former is a VAE for generating in-between fonts (or combining fonts). Its encoder uses a Point net and transformer to compute a code vector of glyph. Its decoder is composed of multiple independent decoders which act on a code vector to reconstruct a point cloud representing a glpyh.

Official PaddlePaddle Implementation of Few-Shot Font Generation by Learning Fine-Grained Local Styles (FsFont)

Official PyTorch implementation of DM-Font (ECCV 2020)

Official PyTorch implementation of LF-Font (Few-shot Font Generation with Localized Style Representations and Factorization) AAAI 2021

Official PyTorch implementation of MX-Font (Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts) ICCV 2021

The unified repository for few-shot font generation methods. This repository includes FUNIT (ICCV'19), DM-Font (ECCV'20), LF-Font (AAAI'21) and MX-Font (ICCV'21).

[AAAI2024] FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning

Open Source Dot Matrix font generator

🔥 [CVPR 2020] STEFANN: Scene Text Editor using Font Adaptive Neural Network (official code).

It is a Python-based tool designed to generate stylish fonts using input images and text. The script processes images, applies transformations, and combines them with text to create visually appealing fonts. It supports various configurations, including custom fonts and rotation options, and outputs the results in organized directories.

Synthesize High-Quality Thai Vector Fonts with DeepVecFont-V2.

基于深度学习的汉字字体补全系统（VQ-VAE+扩散模型）

基於深度學習的漢字字形補全系統

AI-powered Chinese font generation using CycleGAN. An image-to-image translation project to create missing character styles.