Affective Analysis of Professional and Amateur Abstract Paintings Using Statistical Analysis and Art Theory

Sartori, Andreza; Yanulevskaya, Victoria; Salah, Alkım Almila Akdağ; Uijlings, Jasper; Bruni, Elia; Sebe, Nicu

doi:10.1145/2768209

Cited by 38 publications

(38 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Collecting pairwise comparisons has been the method of choice for learning subjective visual attributes such as style, perception, and taste. Examples include learning clothing styles [40], urban appearance [11], emotive responses to GIFs [47], or affective responses to paintings [48]. All these efforts use a two-step process for learning these subjective visual attributes-image ranking, followed by image classification/regression based on the visual attribute.…”

Section: Learning From the Place Pulse 20 Datasetmentioning

confidence: 99%

Deep Learning the City: Quantifying Urban Perception at a Global Scale

Dubey

Naik

Parikh

et al. 2016

Lecture Notes in Computer Science

298

269

View full text Add to dashboard Cite

Abstract. Computer vision methods that quantify the perception of urban environment are increasingly being used to study the relationship between a city's physical appearance and the behavior and health of its residents. Yet, the throughput of current methods is too limited to quantify the perception of cities across the world. To tackle this challenge, we introduce a new crowdsourced dataset containing 110,988 images from 56 cities, and 1,170,000 pairwise comparisons provided by 81,630 online volunteers along six perceptual attributes: safe, lively, boring, wealthy, depressing, and beautiful. Using this data, we train a Siamese-like convolutional neural architecture, which learns from a joint classification and ranking loss, to predict human judgments of pairwise image comparisons. Our results show that crowdsourcing combined with neural networks can produce urban perception data at the global scale.

show abstract

Section: Learning From the Place Pulse 20 Datasetmentioning

confidence: 99%

Deep Learning the City: Quantifying Urban Perception at a Global Scale

Dubey

Naik

Parikh

et al. 2016

Lecture Notes in Computer Science

298

269

View full text Add to dashboard Cite

show abstract

“…Sartori et al [27] used statistical analysis and art theory in a recognition system to find the associated statistical patterns for positive and negative emotions on professional and amateur abstract artworks. For understanding the relationship between artistic principles and emotions, Zhao et al [32] extracted principles-of-art-based emotion features to classify and score image emotions.…”

Section: Abstract Art Evaluationmentioning

confidence: 99%

“…In (27) to (29), we divide A (the above part of image I ) further into three parts. Al is the left part of A, Ar is the right part of A, and Am is the middle part of A.…”

Section: Colormentioning

confidence: 99%

Orientation judgment for abstract paintings

Liu

Dong

Zhang

et al. 2015

Multimed Tools Appl

View full text Add to dashboard Cite

Artists decide the orientation at which an abstract painting should be hung based on their ideas, but the correct orientation is not obvious to other viewers. Some studies have found that abstract paintings at the correct orientations generally get higher aesthetic ratings from viewers. This encourages us to deal with the problem of orientation judgment for abstract paintings through machine learning. First, we design a group of methods to extract features from paintings based on the theories in abstract art. Then a machine leaning framework is proposed using Naive Bayes classifier and BP neural network classifier for training and orientation testing. Experiments show that it can classify abstract paintings into up and non-up ones with performance comparable to human. This is the first work of orientation judgment for abstract paintings through computer simulation, and the results demonstrate the validity of abstract art theories used for feature definition. This work provides a new scheme for exploring the relationship between aesthetic quality of abstract paintings and their computational visual features.

show abstract

“…DeviantArt is an open web community where artists can post their work and interact with others with opinions and ideas; it has more than 200 million art pieces (Sartori et al, 2015).…”

Section: Methodsmentioning

confidence: 99%

Role of gender, professional level, and geographic location of artists on how they represent a story: the case of Little Red Riding Hood

Barrientos¹,

Mónge-Nájera

Barrientos

et al. 2017

URJ

View full text Add to dashboard Cite

Little Red Riding Hood is a widely known classic story and its text has been abundantly analyzed, but no detailed statistical studies have been published about how it has been illustrated. We analyzed 554 images from the public artists' site DeviantArt (January, 2015); classified them according to how the wolf, Little Red, and the environment were represented by the artists; and applied non-parametrical statistical tests to check several hypotheses. When compared with professionals, amateur artists tended to present a more neutral environment, and to humanize the wolf. Female artists were more likely to represent the wolf as a dressed man. Men were more likely to set the story outside of forests, to eroticize Red and to show her confused, scared or uninterested when first meeting the wolf. The neutral attitude of amateurs towards nature suggests indecision, while professional artists seem more used to produce family-friendly images. The female tendency to present the wolf as a man forces them to dress him and may reflect a stronger awareness about the moral of the story, meant to warn young women about men's sexuality. Men deviate more from the forest setting because they feel safer in new environments, and also appear to see Red as a sexually attractive partner and the wolf as a competitor. Artists tended to show no sexual intent between the characters, but those who did were mostly amateurs. The global similarity in art about Little Red Riding Hood indicates that all modern audiences are familiar with the standard representation of the story in books, films and television. This article presents a rigorous quantitative approach to the study of art that can be applied to many other stories and subjects.Key words: artistic interpretation, cultural differences, effect of gender on art, fairy tale representation. RESUMEN: Papel del género, nivel profesional y localización geográfica de los artistas sobre cómo representan una historia: el caso de Caperucita Roja. Caperucita Roja es una historia clásica muy conocida y su texto ha sido abundantemente analizado, pero no se han publicado estudios estadísticos detallados sobre cómo se ha ilustrado. Analizamos 554 imágenes del sitio DeviantArt.com (enero de 2015); clasificándolas de acuerdo a cómo representan los artistas al lobo, caperucita roja, y el ambiente. Aplicamos pruebas estadísticas no paramétricas para evaluar varias hipótesis. En comparación con los profesionales, los artistas aficionados tienden a presentar un ambiente más neutral, y a humanizar al lobo. Los hombres son más propensos a ubicar la historia fuera del bosque, a erotizar a caperucita; y mostrar su confusión, miedo o desinterés durante el primer encuentro con el lobo. La actitud neutral de los aficionados hacia la naturaleza sugiere indecisión, mientras que los profesionales producen imágenes dirigidas a toda la familia. La tendencia femenina a presentar al lobo como un hombre obliga a vestirlo y puede reflejar una mayor conciencia sobre la moral de la historia, destinada a advertir a las...

show abstract

Affective Analysis of Professional and Amateur Abstract Paintings Using Statistical Analysis and Art Theory

Cited by 38 publications

References 38 publications

Deep Learning the City: Quantifying Urban Perception at a Global Scale

Deep Learning the City: Quantifying Urban Perception at a Global Scale

Orientation judgment for abstract paintings

Role of gender, professional level, and geographic location of artists on how they represent a story: the case of Little Red Riding Hood

Contact Info

Product

Resources

About