Confusable Learning for Large-class Few-Shot Classification
This work addresses the problem of large-class few-shot classification for computer vision, offering an incremental improvement by focusing on confusable classes.
The paper tackles the challenge of few-shot image classification with a large number of classes by addressing confusable classes, achieving improved performance over state-of-the-art baselines on datasets like Omniglot, Fungi, and ImageNet.
Few-shot image classification is challenging due to the lack of ample samples in each class. Such a challenge becomes even tougher when the number of classes is very large, i.e., the large-class few-shot scenario. In this novel scenario, existing approaches do not perform well because they ignore confusable classes, namely similar classes that are difficult to distinguish from each other. These classes carry more information. In this paper, we propose a biased learning paradigm called Confusable Learning, which focuses more on confusable classes. Our method can be applied to mainstream meta-learning algorithms. Specifically, our method maintains a dynamically updating confusion matrix, which analyzes confusable classes in the dataset. Such a confusion matrix helps meta learners to emphasize on confusable classes. Comprehensive experiments on Omniglot, Fungi, and ImageNet demonstrate the efficacy of our method over state-of-the-art baselines.