A dual-branch model with inter- and intra-branch contrastive loss for long-tailed recognition

• Published in: Neural networks Vol. 168; pp. 214 - 222
• Main Authors: Chen, Qiong, Huang, Tianlin, Zhu, Geren, Lin, Enlu
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2023
• Subjects: Contrastive learning; Imbalanced learning; Long-tailed recognition; Neural network; Long-tailed recognition; Neural network; Imbalanced learning; Contrastive learning
• ISSN: 0893-6080; 1879-2782
• DOI: 10.1016/j.neunet.2023.09.022

Real-world data often exhibits a long-tailed distribution, in which head classes occupy most of the data, while tail classes only have very few samples. Models trained on long-tailed datasets have poor adaptability to tail classes and the decision boundaries are ambiguous. Therefore, in this paper, we propose a simple yet effective model, named Dual-Branch Long-Tailed Recognition (DB-LTR), which includes an imbalanced learning branch and a Contrastive Learning Branch (CoLB). The imbalanced learning branch, which consists of a shared backbone and a linear classifier, leverages common imbalanced learning approaches to tackle the data imbalance issue. In CoLB, we learn a prototype for each tail class, and calculate an inter-branch contrastive loss, an intra-branch contrastive loss and a metric loss. CoLB can improve the capability of the model in adapting to tail classes and assist the imbalanced learning branch to learn a well-represented feature space and discriminative decision boundary. Extensive experiments on three long-tailed benchmark datasets, i.e., CIFAR100-LT, ImageNet-LT and Places-LT, show that our DB-LTR is competitive and superior to the comparative methods. •An inter-branch contrastive loss is designed to distinguish head- and tail features.•An intra-branch contrastive loss is designed to make the class boundary of tail classes distinct.•A prototype-based metric loss is designed to boost the model’s adaptation ability to tail classes.