Automatic Analysis of Artistic Paintings using Probabilistic and Algorithmic Approaches

Resources

Supplementary Material To The Article.

Average Complexity

Average complexity of each author for different levels of quantization in paintings.

Combining NC and HDC

Combining the NC with the roughness exponent of HDC function

Phylogenetic Tree

Phylogenetic tree of Relationship between authors' fingerprints.

Research Pipeline

Description of the Exposed work

Dataset

4,266 images from 91 well known painters, such as Claude Monet, Frida Kahlo, Henri Matisse, Jackson Pollock, Picasso, Rembrandt, and Salvador Dali.

Source of Dataset: Site , Article
Compute Average Author Complexity

Study average author complexity using NC and NBDM for different quantization levels (4, 6 and 8 bits)
Combining NC and HDC

Study the combination of average author in NC with the roughness exponent α of HDC function regarding stylistic movements
Fingerprint of the authors' complexity.

Compute regional NC (64 Regions) for each of the paintings. Afterwards, compute for each author their average regional complexity.
Phylogenetic Tree

Using each authors' average regional complexity, we computed a distance matrix and used it to create e phylogenetic tree that relates each author.

Average Complexity

Average values of Normalized Compression (NC) and Normalized Block Decomposition method (NBDM) for each author when images were quantitized for 4, 6 and 8 bits.

Combining NC and HDC

Combining the NC with the roughness exponent of HDC function α

Complexity Fingerprint

Compute regional NC (64 Regions) for each of the paintings. Afterwards, compute for each author their average regional complexity.

Albrecht Dürer

Amedeo Modigliani

Andrea Mantegna

Andy Warhol

Arshile Gorky

Camille Corot

Caravaggio

Caspar David Friedrich

Claude Lorrain

Claude Monet

Dante Gabriel Rossetti

David Hockney

Diego Velázquez

Edgar Degas

Édouard Manet

Edvard Munch

Edward Hopper

Egon Schiele

El Lissitzky

Eugène Delacroix

Fernand Léger

Fra Angelico

Francis Bacon

Francisco Goya

Francisco de Zurbarán

Frans Hals

Franz Marc

Frederic Edwin Church

Frida Kahlo

Artemisia Gentileschi

Georges Braque

Georges de La Tour

Georges Seurat

Georgia O'Keeffe

Gerhard Richter

Giorgio de Chirico

Giorgione

Giotto di Bondone

Gustav Klimt

Gustave Courbet

Gustave Moreau

Hans Holbein the Younger

Hans Memling

Henri Matisse

Hieronymus Bosch

Jackson Pollock

Jacques-Louis David

James Ensor

James Abbott McNeill Whistler

Jan van Eyck

Jan Vermeer

Jasper Johns

Jean-François Millet

Jean-Antoine Watteau

Jean Auguste Dominique Ingres

Jean-Michel Basquiat

Joachim Patinir

Joan Miró

John Constable

J. M. W. Turner

Kazimir Malevich

Lucio Fontana

Marc Chagall

Mark Rothko

Max Ernst

Nicolas Poussin

Paul Cézanne

Paul Gauguin

Paul Klee

Peter Paul Rubens

Pablo Picasso

Pierre-Auguste Renoir

Piet Mondrian

Pieter Bruegel the Elder

Raphael

Rembrandt

René Magritte

Rogier van der Weyden

Roy Lichtenstein

Salvador Dalí

Sandro Botticelli

Théodore Géricault

Tintoretto

Titian

Umberto Boccioni

Vincent van Gogh

Wassily Kandinsky

Willem de Kooning

William Blake

William Hogarth

Winslow Homer

Artists Phylogenetic Tree

Phylogenetic tree of Relationship between authors' complexity fingerprints using UPGMA Algorithm.

Phylogenetic tree of Relationship between authors' complexity fingerprints using Kruskal minimum spanning tree Algorithm.

Research Team

Researchers involved in this project.

Jorge Miguel Silva

PhD Student

Diogo Pratas

Researcher

Rui Antunes

PhD Student

Sérgio Matos

Assistant Professor

Armando J. Pinho

Associate Professor

Code of The Project

Funding

This work was funded by National Funds through the FCT in the context of the project UID/CEC/00127/2019 and the research grant SFRH/BD/141851/2018. D.P. is funded by national funds through FCT - Fundação para a Ciência e a Tecnologia, I.P., under the Scientific Employment Stimulus - Institutional Call - CI-CTTI-94-ARH/2019.

Resources

Supplementary Material To The Article.

Average Complexity

Combining NC and HDC

Complexity Fingerprint

Phylogenetic Tree

Research Pipeline

Description of the Exposed work

Dataset

Compute Average Author Complexity

Combining NC and HDC

Fingerprint of the authors' complexity.

Phylogenetic Tree

Average Complexity

Average values of Normalized Compression (NC) and Normalized Block Decomposition method (NBDM) for each author when images were quantitized for 4, 6 and 8 bits.

ANC

NBDM1

NBDM2

Combining NC and HDC

Combining the NC with the roughness exponent of HDC function α

Complexity Fingerprint

Compute regional NC (64 Regions) for each of the paintings. Afterwards, compute for each author their average regional complexity.

Artists Phylogenetic Tree

Phylogenetic tree of Relationship between authors' complexity fingerprints using UPGMA Algorithm.

Phylogenetic tree of Relationship between authors' complexity fingerprints using Kruskal minimum spanning tree Algorithm.

Research Team

Researchers involved in this project.

Jorge Miguel Silva

Diogo Pratas

Rui Antunes

Sérgio Matos

Armando J. Pinho

Funding