OBI designer: zero-shot oracle bone inscription artistic characters generation with multimodal style transfer

Oracle Bone Inscriptions (OBI) artistic character generation combines ancient characters with artificial intelligence to create artworks with cultural heritage and visual impact. Current creation relies on specialized, time-consuming manual skills. We propose OBI-Designer, a zero-shot method to automatically create OBI art characters that relies on pre-trained Vision-Language Models. OBI-Designer uses a two-stage pipeline: glyph synthesis and texture synthesis. The glyph synthesis stage introduces a semantic-driven image vectorization method based on Interval Score Matching (ISM) to optimize character outlines, using additional loss functions to ensure readability. The texture synthesis stage stylizes the glyphs using ControlNet and LoRA fine-tuning. Extensive experiments demonstrate our method’s effectiveness, achieving a Top-1 recognition accuracy of 93.8% and ranking highest in user studies (4.20 out of 5). OBI-Designer further promotes the combination of oracle art synthesis and artificial intelligence.

Authors

Institutions

Publication Details

Journal
npj Heritage Science
Published
2026-03-13
DOI
https://doi.org/10.1038/s40494-026-02417-7
Primary Topic
Generative Adversarial Networks and Image Synthesis
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

OBI designer: zero-shot oracle bone inscription artistic characters generation with multimodal style transfer

Fei Deng, Tao Chen, Li Zhang, Jia Yuan et al.
npj Heritage Science
Generative Adversarial Networks and Image Synthesis
article

OBI designer: zero-shot oracle bone inscription artistic characters generation with multimodal style transfer

Fei Deng, Tao Chen, Li Zhang, Jia Yuan, Guang Long, Chenjun Xu, Ruiyuan Li
article en

Abstract

Oracle Bone Inscriptions (OBI) artistic character generation combines ancient characters with artificial intelligence to create artworks with cultural heritage and visual impact. Current creation relies on specialized, time-consuming manual skills. We propose OBI-Designer, a zero-shot method to automatically create OBI art characters that relies on pre-trained Vision-Language Models. OBI-Designer uses a two-stage pipeline: glyph synthesis and texture synthesis. The glyph synthesis stage introduces a semantic-driven image vectorization method based on Interval Score Matching (ISM) to optimize character outlines, using additional loss functions to ensure readability. The texture synthesis stage stylizes the glyphs using ControlNet and LoRA fine-tuning. Extensive experiments demonstrate our method’s effectiveness, achieving a Top-1 recognition accuracy of 93.8% and ranking highest in user studies (4.20 out of 5). OBI-Designer further promotes the combination of oracle art synthesis and artificial intelligence.

npj Heritage ScienceVol. 14(1)
Southwest University (CN), Hainan University (CN), Hainan Agricultural School (CN)
National Office for Philosophy and Social Sciences
Sustainable cities and communities
Openalex Percentile: Top 23%
Generative Adversarial Networks and Image Synthesis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.