Single-Label Multi-Class Image Classification by Deep Logistic Regression
This work addresses a specific bottleneck in deep learning for image classification, offering incremental improvements over existing methods.
The paper tackles the problem of negative class distraction in logistic regression for single-label multi-class image classification, proposing two novel objective functions that outperform softmax regression in coarse-grained object categorization and fine-grained person identification tasks.
The objective learning formulation is essential for the success of convolutional neural networks. In this work, we analyse thoroughly the standard learning objective functions for multi-class classification CNNs: softmax regression (SR) for single-label scenario and logistic regression (LR) for multi-label scenario. Our analyses lead to an inspiration of exploiting LR for single-label classification learning, and then the disclosing of the negative class distraction problem in LR. To address this problem, we develop two novel LR based objective functions that not only generalise the conventional LR but importantly turn out to be competitive alternatives to SR in single label classification. Extensive comparative evaluations demonstrate the model learning advantages of the proposed LR functions over the commonly adopted SR in single-label coarse-grained object categorisation and cross-class fine-grained person instance identification tasks. We also show the performance superiority of our method on clothing attribute classification in comparison to the vanilla LR function.