Comparing Vision Transformers and Convolutional Neural Networks for Image Classification: A Literature Review
Сравнение Vision Transformers и сверточных нейронных сетей для классификации изображений: обзор литературы
2023-04-28
SCID: 54.1/crqskc3y
Discuss with AI
Convolutional Neural NetworksMulti-Head AttentionVision Transformersdeep learning architecturesimage classification
Figures from the paper
Abstract (AI)
Transformers are models that implement a mechanism of self-attention, individually weighting the importance of each part of the input data. Their use in image classification tasks is still somewhat limited since researchers have so far chosen Convolutional Neural Networks for image classification and transformers were more targeted to Natural Language Processing (NLP) tasks. Therefore, this paper presents a literature review that shows the differences between Vision Transformers (ViT) and Convolutional Neural Networks. The state of the art that used the two architectures for image classification was reviewed and an attempt was made to understand what factors may influence the performance of the two deep learning architectures based on the datasets used, image size, number of target classes (for the classification problems), hardware, and evaluated architectures and top results. The objective of this work is to identify which of the architectures is the best for image classification and under what conditions. This paper also describes the importance of the Multi-Head Attention mechanism for improving the performance of ViT in image classification.
Key Findings
1
It investigates which architectural family performs best under different experimental conditions rather than assuming a universally superior model.
2
Reported top results are examined alongside dataset and implementation factors to identify conditions influencing comparative performance.
3
The analysis emphasizes that Multi-Head Attention is important for improving Vision Transformer performance in image classification.
4
The review addresses Vision Transformers’ relatively limited adoption in image classification compared with CNNs, which have traditionally been preferred.
5
The review compares Vision Transformers and Convolutional Neural Networks for image classification across datasets, image sizes, class counts, hardware, and evaluated architectures.
Research Object
Vision Transformers and Convolutional Neural Networks applied to image classification
Research Subject
Comparative image-classification performance and influencing conditions of Vision Transformers versus Convolutional Neural Networks, including the role of Multi-Head Attention
Publication Details
Publication Date
2023-04-28
Journal
Publisher
ISSN
Cited by
617
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai4
Exploiting Generative AI to Scale up Intelligent Tutoring Systems2023
Literature review as a research methodology: An overview and guidelines2019
Automated Breast Ultrasound Lesions Detection Using Convolutional Neural Networks2017
Transformer Neural Network for Weed and Crop Classification of High Resolution UAV Images2022