mlsanityData to 7 Oct 2026

Total Disentanglement of Font Images into Style and Character Class Features

arXiv 2403.12784 ↗First posted 19 Mar 2024cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

In this paper, we demonstrate a total disentanglement of font images. Total disentanglement is a neural network-based method for decomposing each font image nonlinearly and completely into its style and content (i.e., character class) features. It uses a simple but careful training procedure to extract the common style feature from all `A'-`Z' images in the same font and the common content feature from all `A' (or another class) images in different fonts. These disentangled features guarantee the reconstruction of the original font image. Various experiments have been conducted to understand the performance of total disentanglement. First, it is demonstrated that total disentanglement is achievable with very high accuracy; this is experimental proof of the long-standing open question, ``Does `A'-ness exist?'' Hofstadter (1985). Second, it is demonstrated that the disentangled features produced by total disentanglement apply to a variety of tasks, including font recognition, character recognition, and one-shot font image generation. Code is available here: https://github.com/uchidalab/total_disentanglement

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 49 references that are in the map. First posted on arXiv 19 Mar 2024.

Builds on

32 of 49 references.

References · 32

  1. 2020Exploring Simple Siamese Representation Learningcited by 2,853
  2. 2019DRIT++: Diverse Image-to-Image Translation via Disentangled Representationscited by 244
  3. 2021On Feature Decorrelation in Self-Supervised Learningcited by 160
  4. 2017Multi-Content GAN for Few-Shot Font Style Transfercited by 100
  5. 2019A Learned Representation for Scalable Vector Graphicscited by 91
  6. 2021DG-Font: Deformable Generative Networks for Unsupervised Font Generationcited by 70
  7. 2017Separating Style and Content for Generalized Style Transfercited by 64
  8. 2018Image-to-image translation for cross-domain disentanglementcited by 61
  9. 2020Learning Disentangled Semantic Representation for Domain Adaptationcited by 54
  10. 2019Artistic Glyph Image Synthesis via One-Stage Few-Shot Learningcited by 53
  11. 2023FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learningcited by 48
  12. 2019STEFANN: Scene Text Editor using Font Adaptive Neural Networkcited by 46
20 more
  1. 2020Few-shot Compositional Font Generation with Dual Memorycited by 44
  2. 2022Diff-Font: Diffusion Model for Robust One-Shot Font Generationcited by 40
  3. 2021DRANet: Disentangling Representation and Adaptation Networks for Unsupervised Cross-Domain Adaptationcited by 37
  4. 2018Exploring Disentangled Feature Representation Beyond Face Identificationcited by 36
  5. 2022Few-Shot Font Generation by Learning Fine-Grained Local Stylescited by 34
  6. 2022Look Closer to Supervise Better: One-Shot Font Generation via Component-Based Discriminatorcited by 32
  7. 2017Learning to Write Stylized Chinese Characters by Reading a Handful of Examplescited by 28
  8. 2023DeepVecFont-v2: Exploiting Transformers to Synthesize Vector Fonts with Higher Qualitycited by 27
  9. 2015DeepFont: Identify Your Font from An Imagecited by 24
  10. 2019Large-scale Tag-based Font Retrieval with Generative Feature Learningcited by 22
  11. 2021A Multi-Implicit Neural Representation for Fontscited by 21
  12. 2021ZiGAN: Fine-grained Chinese Calligraphy Font Generation via a Few-shot Style Transfer Approachcited by 20
  13. 2018TET-GAN: Text Effects Transfer via Stylization and Destylizationcited by 19
  14. 2022XMP-Font: Self-Supervised Cross-Modality Pre-training for Few-Shot Font Generationcited by 18
  15. 2019GlyphGAN: Style-Consistent Font Generation Based on Generative Adversarial Networkscited by 16
  16. 2019A Deep Factorization of Style and Structure in Fontscited by 15
  17. 2019TE141K: Artistic Text Benchmark for Text Effect Transfercited by 8
  18. 2021Shared Latent Space of Font Shapes and Their Noisy Impressionscited by 7
  19. 2021Learning Implicit Glyph Shape Representationcited by 7
  20. 2021Scalable Font Reconstruction with Dual Latent Manifoldscited by 4

Built on by

Not cited yet.