Kang Nan, Chang Hong, Ma Bingpeng, Shan Shiguang
IEEE Trans Neural Netw Learn Syst. 2024 Mar;35(3):3437-3449. doi: 10.1109/TNNLS.2022.3192475. Epub 2024 Feb 29.
Data in the visual world often present long-tailed distributions. However, learning high-quality representations and classifiers for imbalanced data is still challenging for data-driven deep learning models. In this work, we aim at improving the feature extractor and classifier for long-tailed recognition via contrastive pretraining and feature normalization, respectively. First, we carefully study the influence of contrastive pretraining under different conditions, showing that current self-supervised pretraining for long-tailed learning is still suboptimal in both performance and speed. We thus propose a new balanced contrastive loss and a fast contrastive initialization scheme to improve previous long-tailed pretraining. Second, based on the motivative analysis on the normalization for classifier, we propose a novel generalized normalization classifier that consists of generalized normalization and grouped learnable scaling. It outperforms traditional inner product classifier as well as cosine classifier. Both the two components proposed can improve recognition ability on tail classes without the expense of head classes. We finally build a unified framework that achieves competitive performance compared with state of the arts on several long-tailed recognition benchmarks and maintains high efficiency.
视觉世界中的数据通常呈现长尾分布。然而，对于数据驱动的深度学习模型而言，为不平衡数据学习高质量的表示和分类器仍然具有挑战性。在这项工作中，我们旨在分别通过对比预训练和特征归一化来改进用于长尾识别的特征提取器和分类器。首先，我们仔细研究了不同条件下对比预训练的影响，表明当前用于长尾学习的自监督预训练在性能和速度方面仍然不是最优的。因此，我们提出了一种新的平衡对比损失和快速对比初始化方案，以改进先前的长尾预训练。其次，基于对分类器归一化的动机分析，我们提出了一种新颖的广义归一化分类器，它由广义归一化和分组可学习缩放组成。它优于传统的内积分类器以及余弦分类器。所提出的两个组件都可以提高对尾部类别的识别能力，而不会以牺牲头部类别的性能为代价。我们最终构建了一个统一的框架，在几个长尾识别基准上与现有技术相比实现了有竞争力的性能，并保持了高效率。
