Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jun 13, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

Ceramic art design generation and aesthetic quality evaluation based on multimodal large language models and

¹College of ART, Hanjiang Normal University, Shiyan, 442000, Hubei, China. zhaowenda2026@163.com.

Scientific Reports

|June 11, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Chiral Memory-Driven Helical Supramolecular Photodetector for Deciphering Circularly Polarized Light.

Advanced materials (Deerfield Beach, Fla.)·2026

Same author

Context-Infused Trajectories: Enhancing Context and Frame Consistency in Reasoning Video Object Segmentation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same author

A self-complementary recombinant adeno-associated virus vector coding for an anchorless prion protein carrying the G127V mutation extends survival in a rodent prion disease model.

bioRxiv : the preprint server for biology·2026

Same author

First-Order Cross-Domain Meta Learning for Few-Shot Remote Sensing Object Classification.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Organic Electrochemical Random Access Memory: From Bio-Inspired to Bio-Integrated.

Advanced materials (Deerfield Beach, Fla.)·2025

Same author

Toward an all-in-one recombinant adeno-associated virus vector for functionally ablating the prion gene using CRISPR-Cas technology.

PloS one·2025

This study introduces an AI framework for ceramic design, merging large language models and diffusion models to generate authentic surface decorations. The AI system enhances cultural authenticity and production efficiency in ceramic art.

Area of Science:

Computer Vision
Artificial Intelligence
Materials Science (Ceramics)

Background:

Traditional ceramic design struggles to balance cultural authenticity with efficient production due to reliance on subjective expertise.
Generating novel ceramic surface decorations often requires extensive manual effort and deep domain knowledge.

Purpose of the Study:

To develop a unified AI framework for generating ceramic surface decorations and evaluating their aesthetic qualities.
To integrate multimodal large language models (MLLMs) and diffusion probabilistic models for enhanced ceramic art design.
To address the limitations of subjective expertise in conventional ceramic design workflows.

Main Methods:

A multimodal semantic parsing module extracts design attributes using instruction-tuned vision-language models and a ceramic knowledge graph.

Keywords:

Ceramic art design Computational aesthetic evaluation Cross-modal semantic alignment Diffusion probabilistic models Multimodal large language models Style-conditioned generation

Related Experiment Videos

Last Updated: Jun 13, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

A style-conditioned diffusion generation network synthesizes designs with fine-grained stylistic control via cross-attention and adaptive layer normalization.

A multi-dimensional aesthetic evaluation model assesses generated designs based on composition, color, texture, and style consistency.

Main Results:

The framework achieved a Fréchet Inception Distance (FID) of 28.35 and an Inception Score (IS) of 10.46 on a dataset of over 21,000 ceramic images.
The aesthetic evaluation model demonstrated high correlation with expert judgment, achieving a Spearman rank correlation coefficient (SRCC) of 0.891 and a Pearson linear correlation coefficient (PLCC) of 0.903.
Ablation studies highlighted the semantic parsing module as critical and multi-dimensional aesthetic evaluation as superior to monolithic scoring.

Conclusions:

The proposed AI framework effectively generates visually appealing ceramic surface decorations while maintaining stylistic control.
The computational aesthetic evaluation provides a reliable proxy for expert judgment, improving design assessment.
This approach offers a novel pathway for enhancing creativity and efficiency in ceramic art design, particularly for traditions like Chinese porcelain and stoneware.