WGAN-CL: A Wasserstein GAN with confidence loss for small-sample augmentation

Jiaqi Mi,Congcong Ma,Lihua Zheng,Man Zhang,Minzan Li,Minjuan Wang

Expert Systems with Applications（2023）

引用 1|浏览3

暂无评分

摘要

•We propose a GAN-based method for image small-sample augmentation named WGAN-CL.•WGAN-CL designs shortcut-stream connections to broaden the model’s solution space.•We design a confidence loss to improve the model's learning capability.•Experiments achieve state-of-the-art performance in image quality and diversity.

查看译文

关键词

Image classification,Plant small-scale dataset,GAN,Data augmentation

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要