A Comprehensive Research of the Development of Classical Convolutional Neural Networks

DOI

10.2991/978-94-6463-540-9_96How to use a DOI?

Keywords

Deep Learning; CNN; Model Architecture

Abstract

Since 2010, with the rapid emergence of deep learning, Convolutional Neural Networks (CNNs) have made significant progress across various domains. In particular, advancements in CNNs have profoundly impacted the field of computer vision, resulting in substantial improvements in tasks such as image classification, object detection, and segmentation. However, as task complexity increases and dataset sizes expand, traditional CNN models face a series of challenges. In response to these obstacles, researchers have devised multiple enhancements and optimization strategies from different perspectives and directions, fostering ongoing developments in structural design and model performance. This paper offers a comprehensive investigation into the evolution of CNNs. The study begins by introducing the standard architecture of CNNs, followed by a delineation of the three significant developmental stages that CNNs have undergone: 1) Traditional Architecture Network, 2) Connectivity-Enhanced Network, and 3) Hybrid Optimization Network. Furthermore, this paper conducts an exhaustive comparison and evaluation of representative models from each stage. Finally, promising directions for CNNs are identified to guide future research endeavors.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Changli Tao
PY  - 2024
DA  - 2024/10/16
TI  - A Comprehensive Research of the Development of Classical Convolutional Neural Networks
BT  - Proceedings of the 2024 2nd International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2024)
PB  - Atlantis Press
SP  - 961
EP  - 969
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-540-9_96
DO  - 10.2991/978-94-6463-540-9_96
ID  - Tao2024
ER  -