Exploring the Impact of Synthetic Data Generation on Texture-based Image Classification Tasks
Yordanov, Borislav and Harvey, Carlo and Williams, Ian and Ashley, Craig and Fairbrass, Paul (2023) Exploring the Impact of Synthetic Data Generation on Texture-based Image Classification Tasks. In: 3rd International Conference on Interactive Media, Smart Systems and Emerging Technologies, 5th - 6th October 2023, Barcelona, Spain.
|
Text
Exploring the Impact of Synthetic Data Generation on Texture-based Image Classification Tasks.pdf - Published Version Available under License Creative Commons Attribution. Download (18MB) |
Abstract
In this study, we introduce a novel pipeline for synthetic data generation of textured surfaces, motivated by the limitations of conventional methods such as Generative Adversarial Networks (GANs) and Computer-Aided Design (CAD) models in our specific context. We also investigate the pipeline's role in an image classification task. The primary objective is to determine the impact of synthetic data generated by our pipeline on classification performance. Using EfficientNetV2-S as our image classifier and a dataset of three texture classes, we find that synthetic data can significantly enhance classification performance when the amount of real data is scarce, corroborating previous research. However, we also observe that the balance between synthetic and real data is crucial, as excessive synthetic data can negatively impact performance when sufficient real data is available. We theorize that this might stem from imperfections in the synthetic data generation process that distort fine details essential for accurate classification, and propose possible improvements to the synthetic data generation pipeline. Furthermore, we acknowledge the potential limitations of our study and provide several promising avenues for future research. This work illuminates the advantages and potential drawbacks of synthetic data in image classification tasks, emphasizing the importance of high-quality, realistic synthetic data that complements, rather than undermines, the use of real data.
| Item Type: | Conference or Workshop Item (Paper) | ||||||
|---|---|---|---|---|---|---|---|
| Identification Number: | https://doi.org/10.2312/imet.20231264 | ||||||
| Dates: |
|
||||||
| Uncontrolled Keywords: | Synthetic Data, Image Classification, Textured Surface | ||||||
| Subjects: | CAH11 - computing > CAH11-01 - computing > CAH11-01-05 - artificial intelligence CAH11 - computing > CAH11-01 - computing > CAH11-01-06 - computer games and animation CAH11 - computing > CAH11-01 - computing > CAH11-01-08 - others in computing |
||||||
| Divisions: | Faculty of Computing, Engineering and the Built Environment Faculty of Computing, Engineering and the Built Environment > School of Computing and Digital Technology |
||||||
| Depositing User: | Borislav Yordanov | ||||||
| Date Deposited: | 26 Oct 2023 13:47 | ||||||
| Last Modified: | 18 Jan 2024 13:54 | ||||||
| URI: | https://www.open-access.bcu.ac.uk/id/eprint/14876 |
Actions (login required)
 |
View Item |
Tools
Tools