CVOct 14, 2021

FocusNet: Classifying Better by Focusing on Confusing Classes

arXiv:2110.07307v314 citations
Originality Incremental advance
AI Analysis

This work addresses the issue of inter-class correlations neglected by one-hot encoding for image classification, representing an incremental improvement over existing methods.

The paper tackles the problem of class confusion in neural networks by proposing a confusion-focusing mechanism with a two-branch architecture and a focus-picking loss function, resulting in improved classification accuracy on common image datasets.

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.

Code Implementations2 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes