2021
DOI: 10.1371/journal.pone.0248414
|View full text |Cite
|
Sign up to set email alerts
|

Compare the performance of the models in art classification

Abstract: Because large numbers of artworks are preserved in museums and galleries, much work must be done to classify these works into genres, styles and artists. Recent technological advancements have enabled an increasing number of artworks to be digitized. Thus, it is necessary to teach computers to analyze (e.g., classify and annotate) art to assist people in performing such tasks. In this study, we tested 7 different models on 3 different datasets under the same experimental setup to compare their art classificati… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

1
13
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
4
1

Relationship

2
8

Authors

Journals

citations
Cited by 23 publications
(14 citation statements)
references
References 36 publications
1
13
0
Order By: Relevance
“…ResNet is the dominant CNN architecture. Traditional CNN architectures have been proven to be extremely powerful for visual arts analysis obtaining state-of-the-art performance in style classification [7,15] and artist attribution tasks [62]. We observe a similar effect as in [15].…”
Section: Fine Art Categorizationsupporting
confidence: 78%
“…ResNet is the dominant CNN architecture. Traditional CNN architectures have been proven to be extremely powerful for visual arts analysis obtaining state-of-the-art performance in style classification [7,15] and artist attribution tasks [62]. We observe a similar effect as in [15].…”
Section: Fine Art Categorizationsupporting
confidence: 78%
“…Various categorization tasks related to images 37 can achieve greater performance with datasets of a limited size with transfer learning than using any other method. Previous work has shown that effective performance can be achieved through pre-trained models fine-tuned on specific tasks 38 , 39 .…”
Section: Related Workmentioning
confidence: 99%
“…The performed analyses span several classification tasks and techniques: from style classification to artist identification, comprising also medium, school, and year classification [ 27 , 28 , 29 ]. These researches are useful to support cultural heritage studies and asset management, e.g., automatic cataloguing of unlabeled works in online and museum collections, but their results can be exploited for more complex applications, such as authentication, stylometry [ 30 ], and forgery detection [ 31 ].…”
Section: Related Workmentioning
confidence: 99%