🤖 AI Summary
Manual assessment of creativity in professional paintings—such as oil paintings—remains inefficient and subjective, lacking scalable, objective evaluation methods. Method: This study introduces the first cross-population CNN model for automatic creativity assessment, designed for both professional painters and children. Unlike prior work focused on doodles or basic shapes, our model leverages a multi-source dataset of 600 high-quality oil paintings and children’s artworks, integrating image classification architecture with built-in interpretability mechanisms to ensure art-critically sound and computationally transparent judgments. Contribution/Results: Experimental evaluation achieves 90% accuracy and accelerates assessment by two orders of magnitude over human raters. This work provides the first empirical validation of deep learning’s efficacy and generalizability in quantifying creativity in professional painting—an advance that fills a critical gap in machine learning applications for high-level visual art assessment.
📝 Abstract
Assessing artistic creativity has long challenged researchers, with traditional methods proving time-consuming. Recent studies have applied machine learning to evaluate creativity in drawings, but not paintings. Our research addresses this gap by developing a CNN model to automatically assess the creativity of human paintings. Using a dataset of six hundred paintings by professionals and children, our model achieved 90% accuracy and faster evaluation times than human raters. This approach demonstrates the potential of machine learning in advancing artistic creativity assessment, offering a more efficient alternative to traditional methods.