Knowledge Commons of Institute of Automation,CAS
Automatically Discovering Novel Visual Categories With Adaptive Prototype Learning | |
Lu Zhang1; Lu Qi2; Xu Yang1; Hong Qiao1; Ming-Hsuan Yang2; Zhiyong Liu1 | |
发表期刊 | https://ieeexplore.ieee.org/abstract/document/10328468 |
ISSN | 0162-8828 |
2024-04 | |
卷号 | 46期号:4页码:2533-2544 |
通讯作者 | Liu, Zhiyong(zhiyong.liu@ia.ac.cn) |
文章类型 | - |
摘要 | This article targets the task of novel category discovery (NCD), which aims to discover unknown categories when a certain number of classes are already known. The NCD task is challenging due to its closeness to real-world scenarios, where we have only encountered some partial classes and corresponding images. Unlike previous approaches to NCD, we propose a novel adaptive prototype learning method that leverages prototypes to emphasize category discrimination and alleviate the issue of missing annotations for novel classes. Concretely, the proposed method consists of two main stages: prototypical representation learning and prototypical self-training. In the first stage, we develop a robust feature extractor that could effectively handle images from both base and novel categories. This ability of instance and category discrimination of the feature extractor is boosted by self-supervised learning and adaptive prototypes. In the second stage, we utilize the prototypes again to rectify offline pseudo labels and train a final parametric classifier for category clustering. We conduct extensive experiments on four benchmark datasets, demonstrating our method’s effectiveness and robustness with state-of-the-art performance. |
关键词 | Image recognition novel category discovery transfer learning |
DOI | 10.1109/TPAMI.2023.3335962 |
收录类别 | SCI |
语种 | 英语 |
资助项目 | NSFC |
项目资助者 | NSFC |
WOS研究方向 | Computer Science ; Engineering |
WOS类目 | Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic |
WOS记录号 | WOS:001180891600036 |
出版者 | IEEE COMPUTER SOC |
七大方向——子方向分类 | 目标检测、跟踪与识别 |
国重实验室规划方向分类 | 视觉信息处理 |
是否有论文关联数据集需要存交 | 否 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/57280 |
专题 | 多模态人工智能系统全国重点实验室 |
通讯作者 | Zhiyong Liu |
作者单位 | 1.State Key Laboratory of Multimodal Artifi cial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences 2.University of California at Merced |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Lu Zhang,Lu Qi,Xu Yang,et al. Automatically Discovering Novel Visual Categories With Adaptive Prototype Learning[J]. https://ieeexplore.ieee.org/abstract/document/10328468,2024,46(4):2533-2544. |
APA | Lu Zhang,Lu Qi,Xu Yang,Hong Qiao,Ming-Hsuan Yang,&Zhiyong Liu.(2024).Automatically Discovering Novel Visual Categories With Adaptive Prototype Learning.https://ieeexplore.ieee.org/abstract/document/10328468,46(4),2533-2544. |
MLA | Lu Zhang,et al."Automatically Discovering Novel Visual Categories With Adaptive Prototype Learning".https://ieeexplore.ieee.org/abstract/document/10328468 46.4(2024):2533-2544. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Zhang 等 - 2024 - Aut(20312KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 |
个性服务 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[Lu Zhang]的文章 |
[Lu Qi]的文章 |
[Xu Yang]的文章 |
百度学术 |
百度学术中相似的文章 |
[Lu Zhang]的文章 |
[Lu Qi]的文章 |
[Xu Yang]的文章 |
必应学术 |
必应学术中相似的文章 |
[Lu Zhang]的文章 |
[Lu Qi]的文章 |
[Xu Yang]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论