Dr Pam Johnston p.johnston2@rgu.ac.uk
Lecturer
Dr Pam Johnston p.johnston2@rgu.ac.uk
Lecturer
Professor Eyad Elyan e.elyan@rgu.ac.uk
Professor
Chrisina Jayne
A collection of Computer Vision application reuse pre-learned features to analyse video frame-by-frame. Those features are classically learned by Convolutional Neural Networks (CNN) trained on high quality images. However, available video content is almost always subject to compression which is nearly never considered during the analysis process. In this paper, we present an empirical study to measure how the visual discrepancy of compressed data limit the learning performance of the CNN model. The learning performance is evaluated using a benchmark of synthetic datasets compressed at various levels using H.264/AVC. We measure the image quality quantitatively using classical evaluation metrics such as Peak Signal to Noise Ratio and Structural SIMilarity. A cross-evaluation is performed to measure the robustness of the CNN model in processing for a wide range of quality-varying visual data. Our experimental results have shown that the performance of the CNN depends on the compression rate. The results show that, in general, higher compression results in lower performance. However performance on lower quality test data can be improved by using lower quality data for CNN training. Finally, our work demonstrates that conditioning the CNN with the compression properties could potentially lead to better learning.
JOHNSTON, P., ELYAN, E. and JAYNE, C. 2018. Spatial effects of video compression on classification in convolutional neural networks. In Proceedings of the 2018 International joint conference on neural networks (IJCNN 2018), 8-13 July 2018, Rio de Janeiro, Brazil. Piscataway, NJ: IEEE [online], article number 8489370. Available from: https://doi.org/10.1109/IJCNN.2018.8489370
Presentation Conference Type | Conference Paper (published) |
---|---|
Conference Name | 2018 International joint conference on neural networks (IJCNN 2018) |
Start Date | Jul 8, 2018 |
End Date | Jul 13, 2018 |
Acceptance Date | Mar 15, 2018 |
Online Publication Date | Jul 8, 2018 |
Publication Date | Dec 31, 2018 |
Deposit Date | Apr 20, 2018 |
Publicly Available Date | Jul 8, 2018 |
Print ISSN | 2161-4393 |
Electronic ISSN | 2161-4407 |
Publisher | Institute of Electrical and Electronics Engineers (IEEE) |
Peer Reviewed | Peer Reviewed |
Article Number | 8489370 |
Series ISSN | 2161-4407 |
DOI | https://doi.org/10.1109/IJCNN.2018.8489370 |
Keywords | CNN; Learning performance; Image quality; Compression |
Public URL | http://hdl.handle.net/10059/2877 |
Contract Date | Apr 20, 2018 |
JOHNSTON 2018 Spatial effects of video compression
(4 Mb)
PDF
Publisher Licence URL
https://creativecommons.org/licenses/by-nc/4.0/
Fusion methods for face presentation attack detection.
(2022)
Journal Article
Deep transfer learning on the aggregated dataset for face presentation attack detection.
(2022)
Journal Article
About OpenAIR@RGU
Administrator e-mail: publications@rgu.ac.uk
This application uses the following open-source libraries:
Apache License Version 2.0 (http://www.apache.org/licenses/)
Apache License Version 2.0 (http://www.apache.org/licenses/)
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2025
Advanced Search