FocusNet: Classifying Better by Focusing on Confusing Classes

TL;DR

FocusNet introduces a two-branch neural network with a focus-picking loss to better distinguish confusing classes, improving image classification accuracy by leveraging inter-class correlations neglected by traditional one-hot encoding.

Contribution

The paper proposes a novel confusion-focusing mechanism and focus-picking loss, enhancing classification by addressing class confusion beyond traditional one-hot labels.

Findings

FocusNet outperforms baseline models on standard datasets.

Focus-picking loss improves accuracy of existing neural networks.

The mechanism effectively reduces class confusion in image classification.

Abstract

Nowadays, most classification networks use one-hot encoding to represent categorical data because of its simplicity. However, one-hot encoding may affect the generalization ability as it neglects inter-class correlations. We observe that, even when a neural network trained with one-hot labels produces incorrect predictions, it still pays attention to the target image region and reveals which classes confuse the network. Inspired by this observation, we propose a confusion-focusing mechanism to address the class-confusion issue. Our confusion-focusing mechanism is implemented by a two-branch network architecture. Its baseline branch generates confusing classes, and its FocusNet branch, whose architecture is flexible, discriminates correct labels from these confusing classes. We also introduce a novel focus-picking loss function to improve classification accuracy by encouraging FocusNet to focus on the most confusing classes. The experimental results validate that our FocusNet is effective for image classification on common datasets, and that our focus-picking loss function can also benefit the current neural networks in improving their classification accuracy.