论文标题
通过域知识作为超级先知的细粒度几乎没有视觉
Fine-Grain Few-Shot Vision via Domain Knowledge as Hyperspherical Priors
论文作者
论文摘要
原型网络已被证明在计算机视觉中的几个学习任务中表现良好。然而，当课程彼此非常相似时（细粒度分类），目前无法考虑先验知识（通过使用表格数据），这些网络很难。使用球形潜在空间来编码原型，我们可以通过最大程度地分开类，同时将域知识纳入信息知识的先验，从而实现很少的细颗粒分类。我们描述了如何构建嵌入A-Priori域信息的原型的超孔，并证明了该方法在挑战基准数据集中进行细粒分类的有效性，并在训练时间内进行了一次性分类和5倍加速的结果。
Prototypical networks have been shown to perform well at few-shot learning tasks in computer vision. Yet these networks struggle when classes are very similar to each other (fine-grain classification) and currently have no way of taking into account prior knowledge (through the use of tabular data). Using a spherical latent space to encode prototypes, we can achieve few-shot fine-grain classification by maximally separating the classes while incorporating domain knowledge as informative priors. We describe how to construct a hypersphere of prototypes that embed a-priori domain information, and demonstrate the effectiveness of the approach on challenging benchmark datasets for fine-grain classification, with top results for one-shot classification and 5x speedups in training time.
