Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillation

TL;DR

This paper introduces a novel two-branch framework utilizing dual ranking statistics and mutual knowledge distillation to improve the discovery of new visual categories from unlabelled images, outperforming existing methods.

Contribution

It proposes a dual-branch learning approach with knowledge transfer techniques for more effective novel category discovery in unlabelled visual data.

Findings

Achieved state-of-the-art results on benchmark datasets.

Effectively leverages local and global features for category discovery.

Demonstrated robustness across various visual recognition tasks.

Abstract

In this paper, we tackle the problem of novel visual category discovery, i.e., grouping unlabelled images from new classes into different semantic partitions by leveraging a labelled dataset that contains images from other different but relevant categories. This is a more realistic and challenging setting than conventional semi-supervised learning. We propose a two-branch learning framework for this problem, with one branch focusing on local part-level information and the other branch focusing on overall characteristics. To transfer knowledge from the labelled data to the unlabelled, we propose using dual ranking statistics on both branches to generate pseudo labels for training on the unlabelled data. We further introduce a mutual knowledge distillation method to allow information exchange and encourage agreement between the two branches for discovering new categories, allowing our model to enjoy the benefits of global and local features. We comprehensively evaluate our method on public benchmarks for generic object classification, as well as the more challenging datasets for fine-grained visual recognition, achieving state-of-the-art performance.