• Stars
    star
    148
  • Rank 249,983 (Top 5 %)
  • Language
    Python
  • License
    Other
  • Created over 3 years ago
  • Updated about 1 year ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

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

MX-Font (ICCV 2021)

NOTICE: We release the unified few-shot font generation repository (clovaai/fewshot-font-generation). If you are interested in using our implementation, please visit the unified repository.

Pytorch implementation of Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Expert. | paper

Song Park1, Sanghyuk Chun2, 3, Junbum Cha3, Bado Lee3, Hyunjung Shim1
1 School of Integrated Technology, Yonsei university
2 NAVER AI Lab
3 NAVER CLOVA

A few-shot font generation (FFG) method has to satisfy two objectives: the generated images should preserve the underlying global structure of the target character and present the diverse local reference style. Existing FFG methods aim to disentangle content and style either by extracting a universal representation style or extracting multiple component-wise style representations. However, previous methods either fail to capture diverse local styles or cannot be generalized to a character with unseen components, e.g., unseen language systems. To mitigate the issues, we propose a novel FFG method, named Multiple Localized Experts Few-shot Font Generation Network (MX-Font). MX-Font extracts multiple style features not explicitly conditioned on component labels, but automatically by multiple experts to represent different local concepts, e.g., left-side sub-glyph. Owing to the multiple experts, MX-Font can capture diverse local concepts and show the generalizability to unseen languages. During training, we utilize component labels as weak supervision to guide each expert to be specialized for different local concepts. We formulate the component assign problem to each expert as the graph matching problem, and solve it by the Hungarian algorithm. We also employ the independence loss and the content-style adversarial loss to impose the content-style disentanglement. In our experiments, MX-Font outperforms previous state-of-the-art FFG methods in the Chinese generation and cross-lingual, e.g., Chinese to Korean, generation.

You can find more related projects on the few-shot font generation at the following links:


Prerequisites

conda install numpy scipy scikit-image tqdm jsonlib-python3 fonttools

Usage

Note that, we only provide the example font files; not the font files used for the training the provided weight (generator.pth). The example font files are downloaded from here.

Preparing Data

  • The examples of datasets are in (./data)

Font files (.ttf)

  • Prepare the TrueType font files(.ttf) to use for the training and the validation.
  • Put the training font files and validation font files into separate directories.

The text files containing the available characters of .ttf files (.txt)

  • If you have the available character list of a .ttf file, save its available characters list to a text file (.txt) with the same name in the same directory with the ttf file.
    • (example) TTF file: data/ttfs/train/MaShanZheng-Regular.ttf, its available characters: data/ttfs/train/MaShanZheng-Regular.txt
  • You can also generate the available characters files automatically using the get_chars_from_ttf.py
# Generating the available characters file

python get_chars_from_ttf.py --root_dir path/to/ttf/dir
  • --root_dir: The root directory to find the .ttf files. All the .ttf files under this directory and its subdirectories will be processed.

The json files with decomposition information (.json)

  • The files for the decomposition information are needed.
    • The files for the Chinese characters are provided. (data/chn_decomposition.json, data/primals.json)
    • If you want to train the model with a language other than Chinese, the files for the decomposition rule (see below) are also needed.
      • Decomposition rule
        • structure: dict (in json format)
        • format: {char: [list of components]}
        • example: {'㐬': ['亠', '厶', '川'], '㐭': ['亠', '囗', '口']}
      • Primals
        • structure: list (in json format)
        • format: [All the components in the decomposition rule file]
        • example: ['亠', '厶', '川', '囗', '口']

Training

Modify the configuration file (cfgs/train.yaml)

- use_ddp:  whether to use DataDistributedParallel, for multi-GPUs.
- port:  the port for the DataDistributedParallel training.

- work_dir:  the directory to save checkpoints, validation images, and the log.
- decomposition:  path to the "decomposition rule" file.
- primals:  path to the "primals" file.

- dset:  (leave blank)
  - train:  (leave blank)
    - data_dir : path to .ttf files for the training
  - val: (leave blank)
    - data_dir : path to .ttf files for the validation
    - source_font : path to .ttf file used as the source font during the validation

Run training

python train.py cfgs/train.yaml
  • arguments
    • path/to/config (first argument): path to configration file.
    • --resume (optional) : path to checkpoint to resume.

Test

Preparing the reference images

  • Prepare the reference images and the .ttf file to use as the source font.
  • The reference images are should be placed in this format:
    * data_dir
    |-- font1
        |-- char1.png
        |-- char2.png
        |-- char3.png
    |-- font2
        |-- char1.png
        |-- char2.png
            .
            .
            .
  • The names of the directory or the image files are not important, however, the images with the same reference style are should be grouped with the same directory.
  • If you want to generate only specific characters, prepare the file containing the list of the characters to generate.
    • The example file is provided. (data/chn_gen.json)

Modify the configuration file (cfgs/eval.yaml)

- dset:  (leave blank)
  - test:  (leave blank)
    - data_dir: path to reference images
    - source_font: path to .ttf file used as the source font during the generation
    - gen_chars_file: path to file of the characters to generate. Leave blank if you want to generate all the available characters in the source font.

Run test

python eval.py \
    cfgs/eval.yaml \
    --weight generator.pth \
    --result_dir path/to/save/images
  • arguments
    • path/to/config (first argument): path to configration file.
    • --weight : path to saved weight to test.
    • --result_dir: path to save generated images.

Code license

This project is distributed under MIT license, except modules.py which is adopted from https://github.com/NVlabs/FUNIT.

MX-Font
Copyright (c) 2021-present NAVER Corp.

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

Acknowledgement

This project is based on clovaai/dmfont and clovaai/lffont.

How to cite

@inproceedings{park2021mxfont,
    title={Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts},
    author={Park, Song and Chun, Sanghyuk and Cha, Junbum and Lee, Bado and Shim, Hyunjung},
    year={2021},
    booktitle={International Conference on Computer Vision (ICCV)},
}

More Repositories

1

donut

Official Implementation of OCR-free Document Understanding Transformer (Donut) and Synthetic Document Generator (SynthDoG), ECCV 2022
Python
5,573
star
2

deep-text-recognition-benchmark

Text recognition (optical character recognition) with deep learning methods, ICCV 2019
Jupyter Notebook
3,692
star
3

stargan-v2

StarGAN v2 - Official PyTorch Implementation (CVPR 2020)
Python
3,478
star
4

CRAFT-pytorch

Official implementation of Character Region Awareness for Text Detection (CRAFT)
Python
3,024
star
5

CutMix-PyTorch

Official Pytorch implementation of CutMix regularizer
Python
1,211
star
6

voxceleb_trainer

In defence of metric learning for speaker recognition
Python
1,029
star
7

WCT2

Software that can perform photorealistic style transfer without the need of any post-processing steps.
Python
869
star
8

synthtiger

Official Implementation of SynthTIGER (Synthetic Text Image Generator), ICDAR 2021
Python
482
star
9

tunit

Rethinking the Truly Unsupervised Image-to-Image Translation - Official PyTorch Implementation (ICCV 2021)
Python
452
star
10

rexnet

Official Pytorch implementation of ReXNet (Rank eXpansion Network) with pretrained models
Python
451
star
11

AdamP

AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights (ICLR 2021)
Python
411
star
12

overhaul-distillation

Official PyTorch implementation of "A Comprehensive Overhaul of Feature Distillation" (ICCV 2019)
Python
409
star
13

cord

CORD: A Consolidated Receipt Dataset for Post-OCR Parsing
384
star
14

cutblur

Rethinking Data Augmentation for Image Super-resolution (CVPR 2020)
Jupyter Notebook
379
star
15

wsolevaluation

Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020)
Python
331
star
16

assembled-cnn

Tensorflow implementation of "Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Network"
Python
329
star
17

generative-evaluation-prdc

Code base for the precision, recall, density, and coverage metrics for generative models. ICML 2020.
Python
239
star
18

ext_portrait_segmentation

Python
238
star
19

ClovaCall

ClovaCall dataset and Pytorch LAS baseline code (Interspeech 2020)
Python
218
star
20

fewshot-font-generation

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).
Python
203
star
21

stargan-v2-tensorflow

StarGAN v2 - Official Tensorflow Implementation (CVPR 2020)
Python
187
star
22

EXTD_Pytorch

Official EXTD Pytorch code
Python
187
star
23

CLEval

CLEval: Character-Level Evaluation for Text Detection and Recognition Tasks
Python
185
star
24

TedEval

TedEval: A Fair Evaluation Metric for Scene Text Detectors
Python
176
star
25

rebias

Official Pytorch implementation of ReBias (Learning De-biased Representations with Biased Representations), ICML 2020
Python
168
star
26

aasist

Official PyTorch implementation of "AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks"
Python
167
star
27

SATRN

Official Tensorflow Implementation of SATRN (CVPR Workshop WTDDLE 2020)
Python
162
star
28

lffont

Official PyTorch implementation of LF-Font (Few-shot Font Generation with Localized Style Representations and Factorization) AAAI 2021
Python
156
star
29

bros

Python
156
star
30

som-dst

SOM-DST: Efficient Dialogue State Tracking by Selectively Overwriting Memory (ACL 2020)
Python
150
star
31

dmfont

Official PyTorch implementation of DM-Font (ECCV 2020)
Python
133
star
32

rainbow-memory

Official pytorch implementation of Rainbow Memory (CVPR 2021)
Python
119
star
33

FocusSeq2Seq

[EMNLP 2019] Mixture Content Selection for Diverse Sequence Generation (Question Generation / Abstractive Summarization)
Python
113
star
34

attention-feature-distillation

Official implementation for (Show, Attend and Distill: Knowledge Distillation via Attention-based Feature Matching, AAAI-2021)
Python
111
star
35

frostnet

FrostNet: Towards Quantization-Aware Network Architecture Search
Python
106
star
36

webvicob

Official Implementation of Web-based Visual Corpus Builder (Webvicob), ICDAR 2023
Python
101
star
37

length-adaptive-transformer

Official Pytorch Implementation of Length-Adaptive Transformer (ACL 2021)
Python
99
star
38

spade

Python
81
star
39

embedding-expansion

Official MXNet implementation of "Embedding Expansion: Augmentation in Embedding Space for Deep Metric Learning" (CVPR 2020)
Python
76
star
40

symmetrical-synthesis

Official Tensorflow implementation of "Symmetrical Synthesis for Deep Metric Learning" (AAAI 2020)
Python
71
star
41

units

Python
70
star
42

lookwhostalking

Look Who’s Talking: Active Speaker Detection in the Wild
Python
70
star
43

subword-qac

Subword Language Model for Query Auto-Completion
Python
67
star
44

ssmix

Official PyTorch Implementation of SSMix (Findings of ACL 2021)
Python
60
star
45

SSUL

[NeurIPS 2021] SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learning
Python
59
star
46

BESTIE

[CVPR 2022] Beyond Semantic to Instance Segmentation: Weakly-Supervised Instance Segmentation via Semantic Knowledge Transfer and Self-Refinement
Python
55
star
47

PointWSSIS

[CVPR2023] The Devil is in the Points: Weakly Semi-Supervised Instance Segmentation via Point-Guided Mask Representation
Python
55
star
48

c3_sinet

Python
52
star
49

puridiver

Official PyTorch Implementation of PuriDivER CVPR 2022.
Python
45
star
50

EResFD

Lightweight Face Detector from CLOVA
Python
44
star
51

minimal-rnr-qa

[NAACL 2021] Designing a Minimal Retrieve-and-Read System for Open-Domain Question Answering
Python
36
star
52

ECLIPSE

(CVPR 2024) ECLIPSE: Efficient Continual Learning in Panoptic Segmentation with Visual Prompt Tuning
Python
34
star
53

group-transformer

Official code for Group-Transformer (Scale down Transformer by Grouping Features for a Lightweight Character-level Language Model, COLING-2020).
Python
25
star
54

ProxyDet

Official implementation of the paper "ProxyDet: Synthesizing Proxy Novel Classes via Classwise Mixup for Open-Vocabulary Object Detection"
Python
22
star
55

GeNAS

Official pytorch implementation for GeNAS: Neural Architecture Search with Better Generalization
Python
15
star
56

meev

Python
12
star
57

pkm-transformers

Official implementation of PKM-augmented language models (Findings of EMNLP 2020)
9
star
58

DCutMix

DCutMix official repo
Python
8
star
59

TVQ-VAE

Official pytorch implementation for TVQ-VAE
Jupyter Notebook
8
star
60

textual-kd-slu

Official Implementation of Textual KD SLU (ICASSP 2021)
Python
6
star
61

vat-d

Official Implementation of VAT-D
Python
5
star
62

ActiveASR_AugCR

Repositoty for Efficient Active Learning for Automatic Speech Recognition via Augmented Consistency Regularization
3
star
63

WSSS-BED

Rethinking Saliency-Guided Weakly-Supervised Semantic Segmentation
Python
1
star