Deep Unbiased Embedding Transfer for Zero-Shot Learning | |
Jia, Zhen1; Zhang, Zhang1; Wang, Liang1,2,3; Shan, Caifeng4; Tan, Tieniu1,2,3 | |
发表期刊 | IEEE TRANSACTIONS ON IMAGE PROCESSING |
ISSN | 1057-7149 |
2020 | |
卷号 | 29期号:29页码:1958-1971 |
摘要 | Zero-shot learning aims to recognize objects which do not appear in the training dataset. Previous prevalent mapping-based zero-shot learning methods suffer from the projection domain shift problem due to the lack of image classes in the training stage. In order to alleviate the projection domain shift problem, a deep unbiased embedding transfer (DUET) model is proposed in this paper. The DUET model is composed of a deep embedding transfer (DET) module and an unseen visual feature generation (UVG) module. In the DET module, a novel combined embedding transfer net which integrates the complementary merits of the linear and nonlinear embedding mapping functions is proposed to connect the visual space and semantic space. Whats more, the end-to-end joint training process is implemented to train the visual feature extractor and the combined embedding transfer net simultaneously. In the UVG module, a visual feature generator trained with a conditional generative adversarial framework is used to synthesize the visual features of the unseen classes to ease the disturbance of the projection domain shift problem. Furthermore, a quantitative index, namely the score of resistance on domain shift (ScoreRDS), is proposed to evaluate different models regarding their resistance capability on the projection domain shift problem. The experiments on five zero-shot learning benchmarks verify the effectiveness of the proposed DUET model. As demonstrated by the qualitative and quantitative analysis, the unseen class visual feature generation, the combined embedding transfer net and the end-to-end joint training process all contribute to alleviating projection domain shift in zero-shot learning. |
关键词 | Visualization Feature extraction Semantics Training Seals Prototypes Indexes Zero-shot learning image classification projection domain shift convolutional neural network generative adversarial network |
DOI | 10.1109/TIP.2019.2947780 |
收录类别 | SCI |
语种 | 英语 |
资助项目 | Natural Science Foundation of China[61633021] ; Natural Science Foundation of China[61525306] ; National Key R&D Program of China[2016YFB1001000] ; Natural Science Foundation of China[61721004] ; Natural Science Foundation of China[61721004] ; National Key R&D Program of China[2016YFB1001000] ; Natural Science Foundation of China[61525306] ; Natural Science Foundation of China[61633021] |
WOS研究方向 | Computer Science ; Engineering |
WOS类目 | Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic |
WOS记录号 | WOS:000501324900023 |
出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
是否为代表性论文 | 是 |
七大方向——子方向分类 | 图像视频处理与分析 |
国重实验室规划方向分类 | 小样本高噪声数据学习 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/29392 |
专题 | 智能感知与计算研究中心 |
通讯作者 | Zhang, Zhang |
作者单位 | 1.Chinese Acad Sci, Natl Lab Pattern Recognit, Ctr Res Intelligent Percept & Comp, Inst Automat, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 3.CAS Ctr Excellence Brain Sci & Intelligence Techn, Beijing 100190, Peoples R China 4.Chinese Acad Sci, AIR, Beijing 100190, Peoples R China |
第一作者单位 | 模式识别国家重点实验室 |
通讯作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Jia, Zhen,Zhang, Zhang,Wang, Liang,et al. Deep Unbiased Embedding Transfer for Zero-Shot Learning[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2020,29(29):1958-1971. |
APA | Jia, Zhen,Zhang, Zhang,Wang, Liang,Shan, Caifeng,&Tan, Tieniu.(2020).Deep Unbiased Embedding Transfer for Zero-Shot Learning.IEEE TRANSACTIONS ON IMAGE PROCESSING,29(29),1958-1971. |
MLA | Jia, Zhen,et al."Deep Unbiased Embedding Transfer for Zero-Shot Learning".IEEE TRANSACTIONS ON IMAGE PROCESSING 29.29(2020):1958-1971. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Deep Unbiased Embedd(3428KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
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