Long-Tailed Learning for Generalized Category Discovery

Abstract

Generalized Category Discovery (GCD) utilizes labeled samples of known classes to discover novel classes in unlabeled samples. Existing methods show effective performance on artificial datasets with balanced distributions. However, real-world datasets are always imbalanced, significantly affecting the effectiveness of these methods. To solve this problem, we propose a novel framework that performs generalized category discovery in long-tailed distributions. We first present a self-guided labeling technique that uses a learnable distribution to generate pseudo-labels, resulting in less biased classifiers. We then introduce a representation balancing process to derive discriminative representations. By mining sample neighborhoods, this process encourages the model to focus more on tail classes. We conduct experiments on public datasets to demonstrate the effectiveness of the proposed framework. The results show that our model exceeds previous state-of-the-art methods.

References in corpus (11)

• SMOTE: Synthetic Minority Over-sampling Technique
• AutoNovel: Automatically Discovering and Learning Novel Visual Categories
• One Million Scenes for Autonomous Driving: ONCE Dataset
• Distribution Aligning Refinery of Pseudo-label for Imbalanced Semi-supervised Learning
• Feature Fusion from Head to Tail for Long-Tailed Visual Recognition
• ABC: Auxiliary Balanced Classifier for Class-imbalanced Semi-supervised Learning
• ProtoGCD: Unified and Unbiased Prototype Learning for Generalized Category Discovery
• Improving Visual Prompt Tuning by Gaussian Neighborhood Minimization for Long-Tailed Visual Recognition
• Novel Class Discovery for Long-tailed Recognition
• Towards Distribution-Agnostic Generalized Category Discovery
• Learning to Discover and Detect Objects