Parametric Classification for Generalized Category Discovery: A Baseline Study

ICCV 2023  ยท  Xin Wen, Bingchen Zhao, Xiaojuan Qi ยท

Generalized Category Discovery (GCD) aims to discover novel categories in unlabelled datasets using knowledge learned from labelled samples. Previous studies argued that parametric classifiers are prone to overfitting to seen categories, and endorsed using a non-parametric classifier formed with semi-supervised k-means. However, in this study, we investigate the failure of parametric classifiers, verify the effectiveness of previous design choices when high-quality supervision is available, and identify unreliable pseudo-labels as a key problem. We demonstrate that two prediction biases exist: the classifier tends to predict seen classes more often, and produces an imbalanced distribution across seen and novel categories. Based on these findings, we propose a simple yet effective parametric classification method that benefits from entropy regularisation, achieves state-of-the-art performance on multiple GCD benchmarks and shows strong robustness to unknown class numbers. We hope the investigation and proposed simple framework can serve as a strong baseline to facilitate future studies in this field. Our code is available at: https://github.com/CVMI-Lab/SimGCD.

PDF Abstract ICCV 2023 PDF ICCV 2023 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Open-World Semi-Supervised Learning CIFAR-10 SimGCD (ViT-B-16) Seen accuracy (50% Labeled) 93.9 # 3
Novel accuracy (50% Labeled) 98.5 # 1
All accuracy (50% Labeled) 97.0 # 1
Open-World Semi-Supervised Learning ImageNet-100 SimGCD (ViT-B-16) Seen accuracy (50% Labeled) 92.4 # 4
Novel accuracy (50% Labeled) 79.1 # 1
All accuracy (50% Labeled) 83.6 # 1

Methods


No methods listed for this paper. Add relevant methods here