PLMCL: Partial-Label Momentum Curriculum Learning for Multi-Label Image Classification

TL;DR

This paper introduces PLMCL, a novel deep learning framework that uses momentum curriculum learning to generate confident pseudo labels for partially labeled and unlabeled images, improving multi-label classification performance.

Contribution

The paper proposes a new partial-label setting and a momentum-based curriculum learning method to effectively utilize unlabeled data and improve classification accuracy.

Findings

PLMCL outperforms existing methods on multiple datasets.

The momentum law stabilizes pseudo label updates, especially early in training.

Adaptive curriculum learning enhances label confidence and model performance.

Abstract

Multi-label image classification aims to predict all possible labels in an image. It is usually formulated as a partial-label learning problem, given the fact that it could be expensive in practice to annotate all labels in every training image. Existing works on partial-label learning focus on the case where each training image is annotated with only a subset of its labels. A special case is to annotate only one positive label in each training image. To further relieve the annotation burden and enhance the performance of the classifier, this paper proposes a new partial-label setting in which only a subset of the training images are labeled, each with only one positive label, while the rest of the training images remain unlabeled. To handle this new setting, we propose an end-to-end deep network, PLMCL (Partial Label Momentum Curriculum Learning), that can learn to produce confident pseudo labels for both partially-labeled and unlabeled training images. The novel momentum-based law updates soft pseudo labels on each training image with the consideration of the updating velocity of pseudo labels, which help avoid trapping to low-confidence local minimum, especially at the early stage of training in lack of both observed labels and confidence on pseudo labels. In addition, we present a confidence-aware scheduler to adaptively perform easy-to-hard learning for different labels. Extensive experiments demonstrate that our proposed PLMCL outperforms many state-of-the-art multi-label classification methods under various partial-label settings on three different datasets.