Computer vision profiling and identification of authentic Jackson Pollock drip paintings
Arts & Communication · 2024
DOI: 10.36922/ac.1628 ↗Kurz gefasst
Fachartikel in der Zeitschrift Arts & Communication, 2024. Verfasst von Lior Shamir. Die Arbeit wurde bisher 2-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
The drip paintings of Jackson Pollock introduced a unique style of abstract expressionism that has attracted substantial interest. Here, a quantitative approach to the analysis of Jackson Pollock’s drip paintings was applied. The analysis was performed using quantitative elements from authentic Jackson Pollock paintings, and comparing them to the same quantitative elements computed from fake Jackson Pollock paintings. Profiling the differences can lead to the identification of unique mathematical elements that characterize Jackson Pollocks artistic style. These elements provide a new way of examining and analyzing the visual content of art. That quantitative analysis can expand on the qualitative analysis methods and provide new insights about visual art. The results show that authentic Jackson Pollock paintings can be differentiated from faked Jackson Pollock paintings in ∼96% of the cases. While fractals provide the strongest differences between authentic and faked Pollock paintings, the analysis also reveals other elements of the paintings that are unique to Jackson Pollock’s style. These elements include textures, entropy, and the distribution of light.
Zitieren
APA
Shamir, L. (2024). Computer vision profiling and identification of authentic Jackson Pollock drip paintings. Arts & Communication. https://doi.org/10.36922/ac.1628
Vancouver
Shamir L. Computer vision profiling and identification of authentic Jackson Pollock drip paintings. Arts & Communication. 2024. doi:10.36922/ac.1628
BibTeX
@article{Shamir2024,
author = {Shamir, Lior},
title = {Computer vision profiling and identification of authentic Jackson Pollock drip paintings},
journal = {Arts & Communication},
year = {2024},
doi = {10.36922/ac.1628}
}
EUVIMED