Abstract

Visual stylometry is a new interdisciplinary research field that sits at the junction of digital humanities, empirical aesthetics, and computer science. Research in this field employs image analysis algorithms to study key aspects of artistic style. The nature of artistic style is the subject of ongoing debate within art history and philosophy of art. Computational and statistical methods in visual stylometry allow researchers to quantify and compare aspects of artistic style over the course of the career of an individual artist, among artists who share in a common artistic style, and across different schools of art. The results of these studies provide clues to the image features that enable viewers to categorize works as belonging to different artistic styles. We are developing a digital image analysis framework for studying paintings called WAIVS (Workflows for Analysis of Images and Visual Stylometry). WAIVS will be powerful enough …

Date

December 8, 2025

Authors

Ricky J Sethi, Catherine A Buell, William P Seeley, Yolanda Gil

Journal

Communications of the ACM, forthcoming