Encyclopedia Enhanced Semantic Embedding for Zero-Shot Learning | |
Jia, Zhen1,2; Zhang, Junge1,2; Huang, Kaiqi1,2,3; Tan, Tieniu1,2,3 | |
2017 | |
会议名称 | International Conference on Image Processing (ICIP) |
会议日期 | 2017 September 17th-20th |
会议地点 | Beijing, China |
摘要 | There are tremendous object categories in the real world besides those in image datasets. Zero-shot learning aims to recognize image categories which are unseen in the training set. A large number of previous zero-shot learning models use word vectors of the class labels directly as category prototypes in the semantic embedding space. But word vectors cannot obtain the global knowledge of an image category sufficiently. In this paper, we propose a new encyclopedia enhanced semantic embedding model to promote the discriminative capability of word vector prototypes with the global knowledge of each image category. The proposed model extracts the TF-IDF key words from encyclopedia articles to acquire the global knowledge of each category. The convex combination of the key words' word vectors acts as the prototypes of the object categories. The prototypes of seen and unseen classes build up the embedding space where the nearest neighbour search is implemented to recognize the unseen images. The experiments show that the proposed method achieves the state-of-the-art performance on the challenging ImageNet Fall 2011 1k2hop dataset. |
关键词 | Zero-shot Learning Image Classification |
收录类别 | EI |
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 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/19662 |
专题 | 智能感知与计算研究中心 |
作者单位 | 1.CRIPAC & NLPR, Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences 3.CAS Center for Excellence in Brain Science and Intelligence Technology |
第一作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Jia, Zhen,Zhang, Junge,Huang, Kaiqi,et al. Encyclopedia Enhanced Semantic Embedding for Zero-Shot Learning[C],2017. |
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 | ||
0001287.pdf(518KB) | 会议论文 | 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