EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Diff-Oracle: Learning Styles and Contents to Augment Realistic Oracle Characters in Diffusion Model

Jing Li, Qiufeng Wang, Siyuan Wang, Rui Zhang, Kaizhu Huang, Erik Cambria

ACM Transactions on Multimedia Computing, Communications, and Applications · 30.06.2026 · 22 (6) · 1-24

DOI: 10.1145/3806389 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Multimedia Computing, Communications, and Applications, online erschienen am 21.05.2026. Verfasst von Jing Li und 5 weiteren Autor:innen. Die Arbeit nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Jiangsu Science and Technology Programme, Open Program of Henan Key Laboratory of Oracle Bone Inscription Information Processing.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recognizing oracle bone scripts plays an important role in Chinese archaeology and philology. However, a significant challenge remains because of the scarcity of oracle character images. To overcome this issue, we propose Diff-Oracle, a novel multi-modal conditional diffusion model that generates a diverse range of controllable oracle characters by inputting random combinations of references. Given the challenge of accurately describing oracle character styles using natural language, Diff-Oracle departs from traditional diffusion models that rely primarily on text prompts by introducing a style encoder. This encoder extracts style prompts from existing oracle character images, where style details are converted into a text embedding format via a pre-trained language-vision model. Additionally, given the lack of explicit content information for oracle characters, ensuring that generated characters accurately represent the intended glyphs is challenging. Therefore, we pre-generate pixel-level paired oracle character images (i.e., style and content images) by an image-to-image translation model, providing content information for the generation process. Meanwhile, Diff-Oracle integrates a content encoder designed to capture specific content details from content reference images. Extensive experiments on Oracle-241 and OBC306 datasets demonstrate that Diff-Oracle significantly outperforms existing generative methods in image quality and diversity. Moreover, Diff-Oracle substantially benefits downstream recognition tasks, outperforming all existing state-of-the-art methods by a large margin. In particular, on the challenging OBC306 dataset, Diff-Oracle achieves a 7.70% accuracy gain in the zero-shot setting and reaches 84.62% accuracy for unseen oracle characters, setting a new benchmark for oracle character recognition. The code is available at https://github.com/JJJingLi/Diff-Oracle .

Förderung

  • National Natural Science Foundation of China
  • Jiangsu Science and Technology Programme
  • Open Program of Henan Key Laboratory of Oracle Bone Inscription Information Processing
  • Foundation of Fujian Key Laboratory of Pattern Recognition and Image Understanding

Zitieren

APA

Li, J., Wang, Q., Wang, S., Zhang, R., Huang, K., & Cambria, E. (2026). Diff-Oracle: Learning Styles and Contents to Augment Realistic Oracle Characters in Diffusion Model. ACM Transactions on Multimedia Computing, Communications, and Applications, 22(6), 1-24. https://doi.org/10.1145/3806389

Vancouver

Li J, Wang Q, Wang S, Zhang R, Huang K, Cambria E. Diff-Oracle: Learning Styles and Contents to Augment Realistic Oracle Characters in Diffusion Model. ACM Transactions on Multimedia Computing, Communications, and Applications. 2026;22(6):1-24. doi:10.1145/3806389

BibTeX
@article{Li2026,
  author = {Li, Jing and Wang, Qiufeng and Wang, Siyuan and Zhang, Rui and Huang, Kaizhu and Cambria, Erik},
  title = {Diff-Oracle: Learning Styles and Contents to Augment Realistic Oracle Characters in Diffusion Model},
  journal = {ACM Transactions on Multimedia Computing, Communications, and Applications},
  year = {2026},
  volume = {22},
  number = {6},
  pages = {1--24},
  doi = {10.1145/3806389}
}
Kostenlos starten →