Tackling Data Bias in Painting Classification with Style Transfer

Mridula Vijendran, Frederick W. B. Li and Hubert P. H. Shum
Proceedings of the 2023 International Conference on Computer Vision Theory and Applications (VISAPP), 2023

Tackling Data Bias in Painting Classification with Style Transfer

Abstract

It is difficult to train classifiers on paintings collections due to model bias from domain gaps and data bias from the uneven distribution of artistic styles. Previous techniques like data distillation, traditional data augmentation and style transfer improve classifier training using task specific training datasets or domain adaptation. We propose a system to handle data bias in small paintings datasets like the Kaokore dataset while simultaneously accounting for domain adaptation in fine-tuning a model trained on real world images. Our system consists of two stages which are style transfer and classification. In the style transfer stage, we generate the stylized training samples per class with uniformly sampled content and style images and train the style transformation network per domain. In the classification stage, we can interpret the effectiveness of the style and content layers at the attention layers when training on the original training dataset and the stylized images. We can tradeoff the model performance and convergence by dynamically varying the proportion of augmented samples in the majority and minority classes. We achieve comparable results to the SOTA with fewer training epochs and a classifier with fewer training parameters.

Downloads

YouTube

Citations

BibTeX

@inproceedings{vijendran23tackling,
 author={Vijendran, Mridula and Li, Frederick W. B. and Shum, Hubert P. H.},
 booktitle={Proceedings of the 2023 International Conference on Computer Vision Theory and Applications},
 series={VISAPP '23},
 title={Tackling Data Bias in Painting Classification with Style Transfer},
 year={2023},
 month={2},
 pages={250--261},
 numpages={12},
 doi={10.5220/0011776600003417},
 issn={2184-4321},
 isbn={978-989-758-634-7},
 publisher={SciTePress},
 location={Lisbon, Portugal},
}

RIS

TY  - CONF
AU  - Vijendran, Mridula
AU  - Li, Frederick W. B.
AU  - Shum, Hubert P. H.
T2  - Proceedings of the 2023 International Conference on Computer Vision Theory and Applications
TI  - Tackling Data Bias in Painting Classification with Style Transfer
PY  - 2023
Y1  - 2 2023
SP  - 250
EP  - 261
DO  - 10.5220/0011776600003417
SN  - 2184-4321
PB  - SciTePress
ER  - 

Plain Text

Mridula Vijendran, Frederick W. B. Li and Hubert P. H. Shum, "Tackling Data Bias in Painting Classification with Style Transfer," in VISAPP '23: Proceedings of the 2023 International Conference on Computer Vision Theory and Applications, pp. 250-261, Lisbon, Portugal, SciTePress, Feb 2023.

Supporting Grants

Similar Research

Mridula Vijendran, Frederick W. B. Li, Jingjing Deng and Hubert P. H. Shum, "ST-SACLF: Style Transfer Informed Self-Attention Classifier for Bias-Aware Painting Classification", Communications in Computer and Information Science (CCIS), 2024
Daniel Organisciak, Edmond S. L. Ho and Hubert P. H. Shum, "Makeup Style Transfer on Low-Quality Images with Weighted Multi-Scale Attention", Proceedings of the 2020 International Conference on Pattern Recognition (ICPR), 2020
Shuang Chen, Amir Atapour-Abarghouei, Edmond S. L. Ho and Hubert P. H. Shum, "INCLG: Inpainting for Non-Cleft Lip Generation with a Multi-Task Image Processing Network", Software Impacts (SIMPAC), 2023

 

 

Last updated on 25 March 2024
RSS Feed