Jingjing Li, Mengmeng Jing, Ke Lu, Lei Zhu, Yang Yang...
TL;DR本文提出了一种新的特征生成网络 AFC-GAN,用于解决基于 GAN 的零样本学习中存在的特征混淆问题,通过提出边界损失和特征混淆得分指标 FCS 等方法可以显著提高模型的性能。
Abstract
Lately, generative adversarial networks (GANs) have been successfully applied
to zero-shot learning (ZSL) and achieved state-of-the-art performance. By
synthesizing virtual unseen visual features, GAN-based metho