Artwork Style Recognition Using Vision Transformers and MLP Mixer
Through the extensive study of transformers, attention mechanisms have emerged as potentially more powerful than sequential recurrent processing and convolution. In this realm, Vision Transformers have gained much research interest, since their architecture changes the dominant paradigm in Computer...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Technologies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7080/10/1/2 |
| _version_ | 1797476329995632640 |
|---|---|
| author | Lazaros Alexios Iliadis Spyridon Nikolaidis Panagiotis Sarigiannidis Shaohua Wan Sotirios K. Goudos |
| author_facet | Lazaros Alexios Iliadis Spyridon Nikolaidis Panagiotis Sarigiannidis Shaohua Wan Sotirios K. Goudos |
| author_sort | Lazaros Alexios Iliadis |
| collection | DOAJ |
| description | Through the extensive study of transformers, attention mechanisms have emerged as potentially more powerful than sequential recurrent processing and convolution. In this realm, Vision Transformers have gained much research interest, since their architecture changes the dominant paradigm in Computer Vision. An interesting and difficult task in this field is the classification of artwork styles, since the artistic style of a painting is a descriptor that captures rich information about the painting. In this paper, two different Deep Learning architectures—Vision Transformer and MLP Mixer (Multi-layer Perceptron Mixer)—are trained from scratch in the task of artwork style recognition, achieving over 39% prediction accuracy for 21 style classes on the WikiArt paintings dataset. In addition, a comparative study between the most common optimizers was conducted obtaining useful information for future studies. |
| first_indexed | 2024-03-09T20:56:20Z |
| format | Article |
| id | doaj.art-529e7dc635434f24b24da82a6fe717cb |
| institution | Directory Open Access Journal |
| issn | 2227-7080 |
| language | English |
| last_indexed | 2024-03-09T20:56:20Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Technologies |
| spelling | doaj.art-529e7dc635434f24b24da82a6fe717cb2023-11-23T22:18:43ZengMDPI AGTechnologies2227-70802021-12-01101210.3390/technologies10010002Artwork Style Recognition Using Vision Transformers and MLP MixerLazaros Alexios Iliadis0Spyridon Nikolaidis1Panagiotis Sarigiannidis2Shaohua Wan3Sotirios K. Goudos4ELEDIA@AUTH, School of Physics, Aristotle University of Thessaloniki, 541 24 Thessaloniki, GreeceELEDIA@AUTH, School of Physics, Aristotle University of Thessaloniki, 541 24 Thessaloniki, GreeceDepartment of Informatics and Telecommunications Engineering, University of Western Macedonia, 501 00 Kozani, GreeceSchool of Information and Safety Engineering, Zhongnan University of Economics and Law, Wuhan 430073, ChinaELEDIA@AUTH, School of Physics, Aristotle University of Thessaloniki, 541 24 Thessaloniki, GreeceThrough the extensive study of transformers, attention mechanisms have emerged as potentially more powerful than sequential recurrent processing and convolution. In this realm, Vision Transformers have gained much research interest, since their architecture changes the dominant paradigm in Computer Vision. An interesting and difficult task in this field is the classification of artwork styles, since the artistic style of a painting is a descriptor that captures rich information about the painting. In this paper, two different Deep Learning architectures—Vision Transformer and MLP Mixer (Multi-layer Perceptron Mixer)—are trained from scratch in the task of artwork style recognition, achieving over 39% prediction accuracy for 21 style classes on the WikiArt paintings dataset. In addition, a comparative study between the most common optimizers was conducted obtaining useful information for future studies.https://www.mdpi.com/2227-7080/10/1/2vision transformerscomputer visiondeep learningartistic style recognition |
| spellingShingle | Lazaros Alexios Iliadis Spyridon Nikolaidis Panagiotis Sarigiannidis Shaohua Wan Sotirios K. Goudos Artwork Style Recognition Using Vision Transformers and MLP Mixer Technologies vision transformers computer vision deep learning artistic style recognition |
| title | Artwork Style Recognition Using Vision Transformers and MLP Mixer |
| title_full | Artwork Style Recognition Using Vision Transformers and MLP Mixer |
| title_fullStr | Artwork Style Recognition Using Vision Transformers and MLP Mixer |
| title_full_unstemmed | Artwork Style Recognition Using Vision Transformers and MLP Mixer |
| title_short | Artwork Style Recognition Using Vision Transformers and MLP Mixer |
| title_sort | artwork style recognition using vision transformers and mlp mixer |
| topic | vision transformers computer vision deep learning artistic style recognition |
| url | https://www.mdpi.com/2227-7080/10/1/2 |
| work_keys_str_mv | AT lazarosalexiosiliadis artworkstylerecognitionusingvisiontransformersandmlpmixer AT spyridonnikolaidis artworkstylerecognitionusingvisiontransformersandmlpmixer AT panagiotissarigiannidis artworkstylerecognitionusingvisiontransformersandmlpmixer AT shaohuawan artworkstylerecognitionusingvisiontransformersandmlpmixer AT sotirioskgoudos artworkstylerecognitionusingvisiontransformersandmlpmixer |