Skip to main content

Research Repository

Advanced Search

ResNet18 performance: Impact of network depth and image resolution on image classification

Liu, Haixia; Brailsford, Tim; Bull, Larry

ResNet18 performance: Impact of network depth and image resolution on image classification Thumbnail


Authors

Profile image of Haixia Liu

Dr Haixia Liu Haixia.Liu@uwe.ac.uk
Senior Lecturer in Computer Science

Lawrence Bull Larry.Bull@uwe.ac.uk
School Director (Research & Enterprise) and Professor



Abstract

This study explores aspects of using ResNet18 for the classification of biomedical images, using the 2D medical images from the MedMNIST library that is widely used for benchmarking. The number of layers within the ResNet18 was studied. The typically used depth (block 4) was generally found to be robust across a variety of datasets at the recommended image resolution (224x224). We found that decreasing the resolution while maintaining block 4 depth can significantly improve performance in some, but not all, cases. The effects of varying both depth and resolution simultaneously were evaluated, and we found a non-linear relationship between depth/resolution and performance. The context of the images seems to be important, with the best performing combinations of network depth and image resolutions varying. We examined the feature maps, and found them to be very variable for the best performing models.

Presentation Conference Type Conference Paper (published)
Conference Name International Conference on Advances in Artificial Intelligence
Start Date Oct 17, 2024
End Date Sep 19, 2024
Acceptance Date Sep 5, 2024
Online Publication Date Mar 3, 2025
Publication Date Mar 3, 2025
Deposit Date Sep 5, 2024
Publicly Available Date Mar 3, 2025
Publisher Association for Computing Machinery (ACM)
Peer Reviewed Peer Reviewed
Pages 351-356
Book Title ICAAI '24: Proceedings of the 2024 8th International Conference on Advances in Artificial Intelligence
ISBN 9798400718014
DOI https://doi.org/10.1145/3704137.3704173
Keywords CCS Concepts, Computing methodologies, Machine learning, Machine learning approaches, Neural networks
Public URL https://uwe-repository.worktribe.com/output/12842105
Publisher URL https://www.acm.org/publications/proceedings

Files






You might also like



Downloadable Citations