mlsanityData to 7 Oct 2026

Font Generation with Missing Impression Labels

arXiv 2203.10348 ↗First posted 19 Mar 2022cs.CVcited by 6

Not on the map. Why?

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.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 24 of its 41 references that are in the map. First posted on arXiv 19 Mar 2022.

Builds on

24 of 41 references.

References · 24

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2022Generative Adversarial Networkscited by 5,603
  3. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  4. 2017Improved Training of Wasserstein GANscited by 2,223
  5. 2014Conditional Generative Adversarial Netscited by 2,152
  6. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  7. 2020Learning from Noisy Labels with Deep Neural Networks: A Surveycited by 580
  8. 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 562
  9. 2018cGANs with Projection Discriminatorcited by 253
  10. 2017Multi-Content GAN for Few-Shot Font Style Transfercited by 100
  11. 2021Multi-Label Learning from Single Positive Labelscited by 81
  12. 2018How good is my GAN?cited by 74
12 more
  1. 2021DG-Font: Deformable Generative Networks for Unsupervised Font Generationcited by 70
  2. 2013Large-scale Multi-label Learning with Missing Labelscited by 58
  3. 2015DeepFont: Identify Your Font from An Imagecited by 24
  4. 2019Large-scale Tag-based Font Retrieval with Generative Feature Learningcited by 22
  5. 2019GlyphGAN: Style-Consistent Font Generation Based on Generative Adversarial Networkscited by 16
  6. 2020On Positive-Unlabeled Classification in GANcited by 14
  7. 2021DeepVecFont: Synthesizing High-quality Vector Fonts via Dual-modality Learningcited by 14
  8. 2020Attribute2Font: Creating Fonts You Want From Attributescited by 12
  9. 2021Impressions2Font: Generating Fonts by Specifying Impressionscited by 10
  10. 2020Let Me Choose: From Verbal Context to Font Selectioncited by 8
  11. 2021Which Parts Determine the Impression of the Font?cited by 7
  12. 2018Class-Distinct and Class-Mutual Image Generation with GANscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 1; 2026: 2.

6 papers.

  1. 2024GRIF-DM: Generation of Rich Impression Fonts using Diffusion Modelscited by 4
  2. 2024Impression-CLIP: Contrastive Shape-Impression Embedding for Fontscited by 3
  3. 2026VecGlypher: Unified Vector Glyph Generation with Language Modelscited by 3
  4. 2025Embedding Font Impression Word Tags Based on Co-occurrencecited by 1
  5. 2024Font Impression Estimation in the Wildcited by 0
  6. 2026Hierarchical Co-Embedding of Font Shapes and Impression Tagscited by 0