Document Type
Article
Author ORCID Identifier
Todd Dobbs https://orcid.org/0000-0001-6711-6962
Abdullah-Al-Raihan Nayeem https://orcid.org/0000-0001-8994-6473
Isaac Cho https://orcid.org/0000-0003-1582-8428
Zbigniew Ras https://orcid.org/0000-0002-8619-914X
Journal/Book Title/Conference
Big Data and Cognitive Computing
Volume
7
Issue
4
Publisher
MDPI AG
Publication Date
10-9-2023
First Page
1
Last Page
12
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Abstract
Art authentication is the process of identifying the artist who created a piece of artwork and is manifested through events of provenance, such as art gallery exhibitions and financial transactions. Art authentication has visual influence via the uniqueness of the artist’s style in contrast to the style of another artist. The significance of this contrast is proportional to the number of artists involved and the degree of uniqueness of an artist’s collection. This visual uniqueness of style can be captured in a mathematical model produced by a machine learning (ML) algorithm on painting images. Art authentication is not always possible as provenance can be obscured or lost through anonymity, forgery, gifting, or theft of artwork. This paper presents an image-only art authentication attribute marker of contemporary art paintings for a very large number of artists. The experiments in this paper demonstrate that it is possible to use ML-generated models to authenticate contemporary art from 2368 to 100 artists with an accuracy of 48.97% to 91.23%, respectively. This is the largest effort for image-only art authentication to date, with respect to the number of artists involved and the accuracy of authentication.
Recommended Citation
Dobbs, T.; Nayeem, A.-A.-R.; Cho, I.; Ras, Z. Contemporary Art Authentication with Large-Scale Classification. Big Data Cogn. Comput. 2023, 7, 162. https://doi.org/10.3390/bdcc7040162
