clovaai/lffont clovaai/lffont

  • Stars
    star
    156
  • Rank 239,589 (Top 5 %)
  • Language
    Python
  • License
    Other
  • Created about 4 years ago
  • Updated 12 months ago
Twitter logo Share LinkedIn logo Share Facebook logo Share
clovaai/lffont
github

clovaai

Reviews

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

Repository Details

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

Few-shot Font Generation with Localized Style Representations and Factorization (AAAI 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.

Official PyTorch implementation of LF-Font | paper | paper (PAMI extension)

Song Park1*, Sanghyuk Chun2*, Junbum Cha2, Bado Lee2, Hyunjung Shim1
* Equal contribution

1 School of Integrated Technology, Yonsei University
2 Clova AI Research, NAVER Corp.

Automatic few-shot font generation is in high demand because manual designs are expensive and sensitive to the expertise of designers. Existing methods of few-shot font generation aims to learn to disentangle the style and content element from a few reference glyphs and mainly focus on a universal style representation for each font style. However, such approach limits the model in representing diverse local styles, and thus make it unsuitable to the most complicated letter system, e.g., Chinese, whose characters consist of a varying number of components (often called "radical") with a highly complex structure. In this paper, we propose a novel font generation method by learning localized styles, namely component-wise style representations, instead of universal styles. The proposed style representations enable us to synthesize complex local details in text designs. However, learning component-wise styles solely from reference glyphs is infeasible in the few-shot font generation scenario, when a target script has a large number of components, e.g., over 200 for Chinese. To reduce the number of reference glyphs, we simplify component-wise styles by a product of component factor and style factor, inspired by low-rank matrix factorization. Thanks to the combination of strong representation and a compact factorization strategy, our method shows remarkably better few-shot font generation results (with only 8 reference glyph images) than other state-of-the-arts, without utilizing strong locality supervision, e.g., location of each component, skeleton, or strokes.

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

Introduction

Pytorch implementation of Few-shot Font Generation with Localized Style Representations and Factorization.

Prerequisites

conda install numpy tqdm lmdb ruamel.yaml jsonlib-python3 msgpack

Usage

Prepare datasets

Build meta file to dump lmdb environment

  • To build a dataset with your own truetype font files (.ttf), a json-format meta file is needed:
    • structure: dict
    • format: {fontname: {"path": path/to/.ttf", "charlist": [chars to dump.]}}
    • example: {"font1": {"path": "./fonts/font1.ttf", "charlist": ["春", "夏", "秋", "冬"]}}

The font file we used as the content font can be accessed here.

Run script

python build_dataset.py \
    --lmdb_path path/to/dump/lmdb \
    --meta_path path/to/meta/file \
    --json_path path/to/save/dict
  • arguments
    • --lmdb_path: path to save lmdb environment.
    • --meta_path: path to meta file of built meta file.
    • --json_path: path to save json file, which contains information of available fonts and characters.
      • saved json file has format like this: {fontname: [saved character list in unicode format]}

Build meta file to train and test

  • train meta (dict, json format)

    • should have keys; "train", "valid", "avail"
    • "train": {font: list of characters} pairs for training, dict
      • key: font name / value: list of chars in the key font.
      • example: {"font1": ["4E00", "4E01"...], "font2": ["4E00", "4E01"...]}
    • "avail": {font: list of characters} pairs which are accessible in lmdb, dict
      • same format with "train"
    • "valid": list of font and list characters for validation, dict
      • should have keys: "seen_fonts", "unseen_fonts", "seen_unis", "unseen_unis"
      • seen fonts(unis) : list of fonts(chars in unicode) in training set.
      • unseen fonts(unis): list of fonts(chars in unicode) not in training set, for validation.
    • An example of train meta file is in meta/train.json.

  • test meta (dict, json format)

    • should have keys; "gen_fonts", "gen_unis", "ref_unis"
    • "gen_fonts": list of fonts to generate.
    • "gen_unis": list of chars to generate, in unicode
    • "ref_unis": list of chars to use as reference chars, in unicode
    • An example of test meta file is in meta/test.json.

Modify the configuration file

We recommend to modify cfgs/custom.yaml rather than cfgs/default.yaml, cfgs/combined.yaml, or cfgs/factorize.yaml.

keys

  • use_half
    • whether to use half tensor. (apex is needed)
  • use_ddp
    • whether to use DataDistributedParallel, for multi-gpus.
  • work_dir
    • the root directory for saved results.
  • data_path
    • path to data lmdb environment.
  • data_meta
    • path to train meta file.
  • content_font
    • the name of font you want to use as source font.
  • other values are hyperparameters for training.

Train

# Phase 1 training
python train.py \
    NAME_phase1 \
    cfgs/custom.yaml cfgs/combined.yaml 

# Phase 2 training
python train.py \
    NAME_phase2 \
    cfgs/custom.yaml cfgs/factorize.yaml \
    --resume ./result/checkpoints/NAME_phase1/800000-NAME_phase1.pth
  • arguments
    • NAME (first argument): name for the experiment.
      • the (checkpoints, validation images, logs) are saved in ./results/(checkpoints, images, logs)/NAME
    • path/to/config (second argument): path to configration file.
      • multiple values are allowed, but their keys should not be repeated.
      • cfgs/combined.yaml : for phase 1 training.
      • cfgs/factorize.yaml: for phase 2 training.
    • --resume (optional) : path to checkpoint to resume.

Test

python evaluator.py \
    cfgs/factorize.yaml \
    --weight weight/generator.pth \
    --img_dir path/to/save/images \
    --test_meta meta/test.json \
    --data_path path/to/data
  • arguments
    • path/to/config (first argument): path to configration file.
    • --weight : path to saved weight to test.
    • --img_dir: path to save generated images.
    • --test_meta: path to test meta file.
    • --data_path: path to lmdb dataset which contatins the reference images.

Code license

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

LF-Font
Copyright (c) 2020-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.

How to cite

@inproceedings{park2021lffont,
    title={Few-shot Font Generation with Localized Style Representations and Factorization},
    author={Park, Song and Chun, Sanghyuk and Cha, Junbum and Lee, Bado and Shim, Hyunjung},
    year={2021},
    booktitle={AAAI Conference on Artificial Intelligence},
}

@article{park2022lffont_extension,
    author={Park, Song and Chun, Sanghyuk and Cha, Junbum and Lee, Bado and Shim, Hyunjung},
    journal = {IEEE Transactions on Pattern Analysis & Machine Intelligence},
    title = {Few-shot Font Generation with Weakly Supervised Localized Representations},
    year = {5555},
    volume = {},
    number = {01},
    issn = {1939-3539},
    pages = {1-17},
    keywords = {},
    doi = {10.1109/TPAMI.2022.3196675},
    publisher = {IEEE Computer Society},
    address = {Los Alamitos, CA, USA},
    month = {aug}
}

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

bros

Python
156
star
29

som-dst

SOM-DST: Efficient Dialogue State Tracking by Selectively Overwriting Memory (ACL 2020)
Python
150
star
30

mxfont

Official PyTorch implementation of MX-Font (Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts) ICCV 2021
Python
148
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