mlsanityData to 7 Oct 2026

Diff-Oracle: Deciphering Oracle Bone Scripts with Controllable Diffusion Model

arXiv 2312.13631 ↗First posted 21 Dec 2023cs.CVcited by 3

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

Deciphering oracle bone scripts plays an important role in Chinese archaeology and philology. However, a significant challenge remains due to the scarcity of oracle character images. To overcome this issue, we propose Diff-Oracle, a novel approach based on diffusion models to generate a diverse range of controllable oracle characters. Unlike traditional diffusion models that operate primarily on text prompts, Diff-Oracle incorporates a style encoder that utilizes style reference images to control the generation style. This encoder extracts style prompts from existing oracle character images, where style details are converted into a text embedding format via a pretrained language-vision model. On the other hand, a content encoder is integrated within Diff-Oracle to capture specific content details from content reference images, ensuring that the generated characters accurately represent the intended glyphs. To effectively train Diff-Oracle, we pre-generate pixel-level paired oracle character images (i.e., style and content images) by an image-to-image translation model. Extensive qualitative and quantitative experiments are conducted on datasets Oracle-241 and OBC306. While significantly surpassing present generative methods in terms of image generation, Diff-Oracle substantially benefits downstream oracle character recognition, outperforming all existing SOTAs by a large margin. In particular, on the challenging OBC306 dataset, Diff-Oracle leads to an accuracy gain of 7.70% in the zero-shot setting and is able to recognize unseen oracle character images with the accuracy of 84.62%, achieving a new benchmark for deciphering oracle bone scripts.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 43 of its 66 references that are in the map. First posted on arXiv 21 Dec 2023.

Builds on

43 of 66 references.

Closest ancestors · 7

Same direction, newest first

  1. 2023Zero-shot Generation of Training Data with Denoising Diffusion Probabilistic Model for Handwritten Chinese Character Recognition25 May
  2. 2023Adding Conditional Control to Text-to-Image Diffusion Models10 Feb
  3. 2023GLIGEN: Open-Set Grounded Text-to-Image Generation17 Jan
  4. 2022Diff-Font: Diffusion Model for Robust One-Shot Font Generation12 Dec
  5. 2022Inversion-Based Style Transfer with Diffusion Models23 Nov
  6. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding23 May
  7. 2021Palette: Image-to-Image Diffusion Models10 Nov

Other references · 36

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  5. 2020Denoising Diffusion Implicit Modelscited by 9,655
  6. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  7. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  8. 2016Densely Connected Convolutional Networkscited by 6,691
  9. 2022Classifier-Free Diffusion Guidancecited by 5,717
  10. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  11. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  12. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,231
24 more
  1. 2014Conditional Generative Adversarial Netscited by 2,152
  2. 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,509
  3. 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
  4. 2020Contrastive Learning for Unpaired Image-to-Image Translationcited by 637
  5. 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 562
  6. 2022EGSDE: Unpaired Image-to-Image Translation via Energy-Guided Stochastic Differential Equationscited by 172
  7. 2020Remix: Rebalanced Mixupcited by 144
  8. 2021UNIT-DDPM: UNpaired Image Translation with Denoising Diffusion Probabilistic Modelscited by 133
  9. 2021Inverse design of 3d molecular structures with conditional generative neural networkscited by 93
  10. 2019Generative Adversarial Minority Oversamplingcited by 72
  11. 2021DG-Font: Deformable Generative Networks for Unsupervised Font Generationcited by 70
  12. 2020Few-shot Compositional Font Generation with Dual Memorycited by 44
  13. 2020Unpaired Image-to-Image Translation using Adversarial Consistency Losscited by 43
  14. 2022Exploring Patch-wise Semantic Relation for Contrastive Learning in Image-to-Image Translation Taskscited by 42
  15. 2022Few-Shot Font Generation by Learning Fine-Grained Local Stylescited by 34
  16. 2023Fill-Up: Balancing Long-Tailed Data with Generative Modelscited by 34
  17. 2017Auto-Encoder Guided GAN for Chinese Calligraphy Synthesiscited by 29
  18. 2023CF-Font: Content Fusion for Few-shot Font Generationcited by 26
  19. 2023AGTGAN: Unpaired Image Translation for Photographic Ancient Character Generationcited by 17
  20. 2022Unsupervised Structure-Texture Separation Network for Oracle Character Recognitioncited by 15
  21. 2022Improving GANs for Long-Tailed Data through Group Spectral Regularizationcited by 14
  22. 2020PatchUp: A Feature-Space Block-Level Regularization Technique for Convolutional Neural Networkscited by 12
  23. 2019Image-to-Image Translation with Multi-Path Consistency Regularizationcited by 6
  24. 2022Exploring Negatives in Contrastive Learning for Unpaired Image-to-Image Translationcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 2.

3 papers.

  1. 2024OracleSage: Towards Unified Visual-Linguistic Understanding of Oracle Bone Scripts through Cross-Modal Knowledge Fusioncited by 10
  2. 2026Towards Computational Chinese Paleographycited by 0
  3. 2026Enhancing Oracle Bone Inscription Recognition via Multi-Scale Layer Attentioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025OracleFusion: Assisting the Decipherment of Oracle Bone Script with Structurally Constrained Semantic Typography
  2. 2024Deciphering Oracle Bone Language with Diffusion Models
  3. 2025Mitigating Long-tail Distribution in Oracle Bone Inscriptions: Dataset, Model, and Benchmark
  4. 2026FROD: Feature Matching Residual Denoising Oracle Bone Decipher
  5. 2026Specializing Large Models for Oracle Bone Script Interpretation via Component-Grounded Multimodal Knowledge Augmentation
  6. 2025Interpretable Oracle Bone Script Decipherment through Radical and Pictographic Analysis with LVLMs
  7. 2026AlphaOracle: Oracle bone script decipherment via human-workflow-inspired deep learning
  8. 2025PictOBI-20k: Unveiling Large Multimodal Models in Visual Decipherment for Pictographic Oracle Bone Characters