# Font Generation with Missing Impression Labels

arXiv: https://arxiv.org/abs/2203.10348 · HTML: https://mlsanity.com/paper/2203.10348/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 41 references that are in the map.
- First posted on arXiv 19 Mar 2022.

## Builds on (most cited first, 24 of 24)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Conditional Generative Adversarial Nets](https://mlsanity.com/paper/1411.1784/) (2014)
- [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/1511.06434/) (2015)
- [Learning from Noisy Labels with Deep Neural Networks: A Survey](https://mlsanity.com/paper/2007.08199/) (2020)
- [Conditional Image Synthesis With Auxiliary Classifier GANs](https://mlsanity.com/paper/1610.09585/) (2016)
- [cGANs with Projection Discriminator](https://mlsanity.com/paper/1802.05637/) (2018)
- [Multi-Content GAN for Few-Shot Font Style Transfer](https://mlsanity.com/paper/1712.00516/) (2017)
- [Multi-Label Learning from Single Positive Labels](https://mlsanity.com/paper/2106.09708/) (2021)
- [How good is my GAN?](https://mlsanity.com/paper/1807.09499/) (2018)
- [DG-Font: Deformable Generative Networks for Unsupervised Font Generation](https://mlsanity.com/paper/2104.03064/) (2021)
- [Large-scale Multi-label Learning with Missing Labels](https://mlsanity.com/paper/1307.5101/) (2013)
- [DeepFont: Identify Your Font from An Image](https://mlsanity.com/paper/1507.03196/) (2015)
- [Large-scale Tag-based Font Retrieval with Generative Feature Learning](https://mlsanity.com/paper/1909.02072/) (2019)
- [GlyphGAN: Style-Consistent Font Generation Based on Generative Adversarial Networks](https://mlsanity.com/paper/1905.12502/) (2019)
- [DeepVecFont: Synthesizing High-quality Vector Fonts via Dual-modality Learning](https://mlsanity.com/paper/2110.06688/) (2021)
- [On Positive-Unlabeled Classification in GAN](https://mlsanity.com/paper/2002.01136/) (2020)
- [Attribute2Font: Creating Fonts You Want From Attributes](https://mlsanity.com/paper/2005.07865/) (2020)
- [Impressions2Font: Generating Fonts by Specifying Impressions](https://mlsanity.com/paper/2103.10036/) (2021)
- [Let Me Choose: From Verbal Context to Font Selection](https://mlsanity.com/paper/2005.01151/) (2020)
- [Which Parts Determine the Impression of the Font?](https://mlsanity.com/paper/2103.14216/) (2021)
- [Class-Distinct and Class-Mutual Image Generation with GANs](https://mlsanity.com/paper/1811.11163/) (2018)

## Built on by (most cited first, 6 of 6)

- [GRIF-DM: Generation of Rich Impression Fonts using Diffusion Models](https://mlsanity.com/paper/2408.07259/) (2024)
- [Impression-CLIP: Contrastive Shape-Impression Embedding for Fonts](https://mlsanity.com/paper/2402.16350/) (2024)
- [VecGlypher: Unified Vector Glyph Generation with Language Models](https://mlsanity.com/paper/2602.21461/) (2026)
- [Embedding Font Impression Word Tags Based on Co-occurrence](https://mlsanity.com/paper/2508.18825/) (2025)
- [Font Impression Estimation in the Wild](https://mlsanity.com/paper/2402.15236/) (2024)
- [Hierarchical Co-Embedding of Font Shapes and Impression Tags](https://mlsanity.com/paper/2604.04158/) (2026)

## Abstract (arXiv)

Our goal is to generate fonts with specific impressions, by training a generative adversarial network with a font dataset with impression labels. The main difficulty is that font impression is ambiguous and the absence of an impression label does not always mean that the font does not have the impression. This paper proposes a font generation model that is robust against missing impression labels. The key ideas of the proposed method are (1)a co-occurrence-based missing label estimator and (2)an impression label space compressor. The first is to interpolate missing impression labels based on the co-occurrence of labels in the dataset and use them for training the model as completed label conditions. The second is an encoder-decoder module to compress the high-dimensional impression space into low-dimensional. We proved that the proposed model generates high-quality font images using multi-label data with missing labels through qualitative and quantitative evaluations.