期刊
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
卷 44, 期 10, 页码 6767-6781出版社
IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2021.3091944
关键词
Task analysis; Ranking (statistics); Data models; Annotations; Benchmark testing; Visualization; Transfer learning; Novel category discovery; deep transfer clustering; clustering; classification; incremental learning
资金
- EPSRC [Seebibyte EP/M013774/1, Mathworks/DTA DFR02620, ERC IDIU-638009]
- Nielsen
In this paper, a method called AutoNovel is proposed to discover novel classes in an image collection. By utilizing self-supervised learning, ranking statistics, and optimizing a joint objective function, both supervised classification of labeled data and clustering of unlabeled data are improved. Furthermore, a method to estimate the number of new classes is also proposed.
We tackle the problem of discovering novel classes in an image collection given labelled examples of other classes. We present a new approach called AutoNovel to address this problem by combining three ideas: (1) we suggest that the common approach of bootstrapping an image representation using the labelled data only introduces an unwanted bias, and that this can be avoided by using self-supervised learning to train the representation from scratch on the union of labelled and unlabelled data; (2) we use ranking statistics to transfer the model's knowledge of the labelled classes to the problem of clustering the unlabelled images; and, (3) we train the data representation by optimizing a joint objective function on the labelled and unlabelled subsets of the data, improving both the supervised classification of the labelled data, and the clustering of the unlabelled data. Moreover, we propose a method to estimate the number of classes for the case where the number of new categories is not known a priori. We evaluate AutoNovel on standard classification benchmarks and substantially outperform current methods for novel category discovery. In addition, we also show that AutoNovel can be used for fully unsupervised image clustering, achieving promising results.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据