Computational identification of significant actors in paintings through symbols and attributes

Stork, David G.; Bourached, Anthony; Cann, George; Griffiths, Ryan‐Rhys

doi:10.2352/issn.2470-1173.2021.14.cvaa-015

Cited by 7 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We believe the methods presented in this paper should find use in computational approaches to art scholarship, particularly higher-level interpretation tasks such as extracting simple meanings from artworks. [17] Our technique of training neural networks with surrogate photographs is not limited to the task of semantic segmentation and could facilitate the transfer of image analysis tools that already exist for natural photographs to the domain fine art paintings. Conceivable examples include art conservation and analysis through digital inpainting of paintings based on deep networks trained with surrogate artworks from the relevant style.…”

Section: Discussionmentioning

confidence: 99%

“…Currently the most accurate segmentation methods rely on deep neural networks trained with large corpora of photographs or stills from videos-up to hundreds of thousands of examples. [12,2] Segmentation is also a key early step in many techniques of automatic art analysis, including the analysis of compositional styles, identifying figures or "actors" as a step to inferring the story, moral, or meaning expressed by an artwork, [17] and others. For a number of reasons, semantic image segmentation of art images has proven difficult for machine learning methods based on deep networks trained with photographs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improving semantic segmentation of fine art images using photographs rendered in a style learned from artworks

Thomas¹,

G.²

2022

Self Cite

View full text Add to dashboard Cite

Our central goal was to create automatic methods for semantic segmentation of human figures in images of fine art paintings. This is a difficult problem because the visual properties and statistics of artwork differ markedly from the natural photographs widely used in research in automatic segmentation. We used a deep neural network to transfer artistic style from paintings across several centuries to modern natural photographs in order to create a large data set of surrogate art images. We then used this data set to train a separate deep network for semantic image segmentation of genuine art images. Such data augmentation led to great improvement in the segmentation of difficult genuine artworks, revealed both qualitatively and quantitatively. Our unique technique of creating surrogate artworks should find wide use in many tasks in the growing field of computational analysis of fine art.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Improving semantic segmentation of fine art images using photographs rendered in a style learned from artworks

Thomas¹,

G.²

2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Since the 2010s, there has been an explosion in the number of papers exploring art forms using computational methods ([ 5 , 6 , 7 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 ]). This has been due to the more formal organisation of image-based datasets, and the large-scale use of neural-network- and data-science-based approaches for image analysis.…”

Section: Computation and Artmentioning

confidence: 99%

“…Another domain of inquiry is aesthetics , where methods to detect lines, contours, and signs are identified to measure aesthetic content ([ 15 , 16 ]). Using both deep-learning and traditional approaches, identification of such symbols and signs, as well as actors, goes towards building semantic interpretations of paintings [ 17 ].…”

Section: Computation and Artmentioning

confidence: 99%

WikiArtVectors: Style and Color Representations of Artworks for Cultural Analysis via Information Theoretic Measures

Desikan

Shimao

Miton

2022

Entropy

View full text Add to dashboard Cite

With the increase in massive digitized datasets of cultural artefacts, social and cultural scientists have an unprecedented opportunity for the discovery and expansion of cultural theory. The WikiArt dataset is one such example, with over 250,000 high quality images of historically significant artworks by over 3000 artists, ranging from the 15th century to the present day; it is a rich source for the potential mining of patterns and differences among artists, genres, and styles. However, such datasets are often difficult to analyse and use for answering complex questions of cultural evolution and divergence because of their raw formats as image files, which are represented as multi-dimensional tensors/matrices. Recent developments in machine learning, multi-modal data analysis and image processing, however, open the door for us to create representations of images that extract important, domain-specific features from images. Art historians have long emphasised the importance of art style, and the colors used in art, as ways to characterise and retrieve art across genre, style, and artist. In this paper, we release a massive vector-based dataset of paintings (WikiArtVectors), with style representations and color distributions, which provides cultural and social scientists with a framework and database to explore relationships across these two vital dimensions. We use state-of-the-art deep learning and human perceptual color distributions to extract the representations for each painting, and aggregate them across artist, style, and genre. These vector representations and distributions can then be used in tandem with information-theoretic and distance metrics to identify large-scale patterns across art style, genre, and artist. We demonstrate the consistency of these vectors, and provide early explorations, while detailing future work and directions. All of our data and code is publicly available on GitHub.

show abstract

“…Our work presented here is inspired by recent work using deep networks that address a related problem: identifying key figures or "actors" in paintings as a step to understanding a work's meaning. 4 That previous approach automatically identified and located key signs or iconographic attributes in an artwork, and found the closest segmented human figure. Thus, for instance, a segmented human figure would be automatically identified as St. John the Baptist because of its proximity to his iconographic attribute, the crucifixion cross; likewise Christ was identified by his proximity to one of his attributes, a dove; and so on.…”

Section: Introduction: Extracting Meanings Conveyed Through Imagesmentioning

confidence: 99%

Extracting associations and meanings of objects depicted in artworks through bi-modal deep networks

Gregory¹,

Ryan-Rhys²,

Anthony³

et al. 2022

Self Cite

View full text Add to dashboard Cite

We present a novel bi-modal system based on deep networks to address the problem of learning associations and simple meanings of objects depicted in "authored" images, such as fine art paintings and drawings. Our overall system processes both the images and associated texts in order to learn associations between images of individual objects, their identities and the abstract meanings they signify. Unlike past deep net that describe depicted objects and infer predicates, our system identifies meaning-bearing objects ("signifiers") and their associations ("signifieds") as well as basic overall meanings for target artworks. Our system had precision of 48% and recall of 78% with an F1 metric of 0.6 on a curated set of Dutch vanitas paintings, a genre celebrated for its concentration on conveying a meaning of great import at the time of their execution. We developed and tested our system on fine art paintings but our general methods can be applied to other authored images.

show abstract

Computational identification of significant actors in paintings through symbols and attributes

Abstract: Fast track article for IS&T International Symposium on Electronic Imaging 2021: Computer Vision and Image Analysis of Art 2021 proceedings.

Cited by 7 publications

References 4 publications

Improving semantic segmentation of fine art images using photographs rendered in a style learned from artworks

Improving semantic segmentation of fine art images using photographs rendered in a style learned from artworks

WikiArtVectors: Style and Color Representations of Artworks for Cultural Analysis via Information Theoretic Measures

Extracting associations and meanings of objects depicted in artworks through bi-modal deep networks

Contact Info

Product

Resources

About