Spherical Zero-Shot Learning
Spherical Zero-Shot Learning
复制标题
球形零样本学习
DOI：
10.1109/tcsvt.2021.3067067
复制
发表时间：
2022-02
影响因子：
8.4
通讯作者：
Lei Zhang
中科院分区：
文献类型：
--
作者：
Jiayi Shen;Zehao Xiao;Xiantong Zhen;Lei Zhang
零镜头学习(ZSL)是一项从看得见的课程到看不见的课程的高度非平凡的任务。在本文中，我们提出了球面零镜头学习(SZSL)来解决ZSL中的主要挑战。通过将球面嵌入空间中的相似性度量分解为半径和角度，SZSL可以将类映射到不同半径的超球面上，从而大大增加了SZSL的灵活性。具体地说，我们引入了角度上的球面对齐来尽可能均匀地扩展类，以缓解Hubness问题，同时保留类间的语义结构，使对齐更加合理。我们还引入了基于最小熵的正则化的球面校正方法，对不可见类采用比可见类更大的半径，以减少预测偏差。在五个中等规模的基准测试和大规模的ImageNet数据集上的大量实验表明，该方法在ZSL的传统和通用设置下一致获得了优越的性能。
Zero-shot Learning (ZSL) is a highly non-trivial task to generalize from seen to unseen classes. In this paper, we propose spherical zero-shot learning (SZSL) to address the major challenges in ZSL. By decoupling the similarity metric in the spherical embedding space into radius and angle, our SZSL can map classes to hyperspherical surfaces of different radiuses, which greatly increases its flexibility. Specifically, we introduce the spherical alignment on angles to spread classes as uniformly as possible to alleviate the hubness problem and simultaneously preserve the inter-class semantic structure to make the alignment more reasonable. We also introduce the spherical calibration with a minimum entropy based regularizer by adopting a larger radius for unseen classes than seen classes to reduce the prediction bias. Extensive experiments on five middle-scale benchmarks and large-scale ImageNet dataset demonstrate that the proposed approach consistently achieves superior performance for the traditional and generalized settings of ZSL.