TY - GEN
T1 - Data Augmentation Framework for Improving Image Recognition Using cycleGAN
AU - Kim, Sangmin
AU - Deribe, Selome Tesfaye
AU - Byun, Gyurin
AU - Joo, Kyeongjin
AU - Pyo, Jeongwon
AU - Kuc, Taeyong
AU - Choo, Hyunseung
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - This paper proposes a novel framework that significantly enhances the performance of semantic segmentation models in recognizing specific objects. Leveraging the capabilities of Generative Adversarial Networks (GANs), particularly Cycle-GAN, this framework focuses on augmenting high-quality data to improve object recognition in autonomous driving and other applications. The study utilizes a dataset of 5,005 road images, enriched with polygon labels for precise object recognition. Key advancements in this research include the implementation of feature matching and fact forcing techniques to stabilize and integrate GAN performance, thereby overcoming common challenges like mode collapse, slow training, and overfitting. In the performance-enhanced GAN model, we improved the Discriminator Loss from the original 1.0517 to 0.0001, achieving convergence to zero 66.67% faster.
AB - This paper proposes a novel framework that significantly enhances the performance of semantic segmentation models in recognizing specific objects. Leveraging the capabilities of Generative Adversarial Networks (GANs), particularly Cycle-GAN, this framework focuses on augmenting high-quality data to improve object recognition in autonomous driving and other applications. The study utilizes a dataset of 5,005 road images, enriched with polygon labels for precise object recognition. Key advancements in this research include the implementation of feature matching and fact forcing techniques to stabilize and integrate GAN performance, thereby overcoming common challenges like mode collapse, slow training, and overfitting. In the performance-enhanced GAN model, we improved the Discriminator Loss from the original 1.0517 to 0.0001, achieving convergence to zero 66.67% faster.
KW - Computer Vision
KW - Generative Adversarial Networks
KW - Optimization modelling
KW - Semantic Segmentation
UR - https://www.scopus.com/pages/publications/85186144638
U2 - 10.1109/IMCOM60618.2024.10418294
DO - 10.1109/IMCOM60618.2024.10418294
M3 - Conference contribution
AN - SCOPUS:85186144638
T3 - Proceedings of the 2024 18th International Conference on Ubiquitous Information Management and Communication, IMCOM 2024
BT - Proceedings of the 2024 18th International Conference on Ubiquitous Information Management and Communication, IMCOM 2024
A2 - Lee, Sukhan
A2 - Choo, Hyunseung
A2 - Ismail, Roslan
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 18th International Conference on Ubiquitous Information Management and Communication, IMCOM 2024
Y2 - 3 January 2024 through 5 January 2024
ER -