Knowledge Commons of Institute of Automation,CAS
AMVH: Asymmetric Multi-Valued Hashing | |
Da, Cheng1,2![]() ![]() ![]() ![]() ![]() ![]() | |
2017-07-30 | |
会议名称 | IEEE Computer Vision and Pattern Recognition (CVPR) |
会议日期 | 2017-07-22 |
会议地点 | Honolulu |
摘要 | Most existing hashing methods resort to binary codesfor similarity search, owing to the high efficiency of computation and storage. However, binary codes lack enough capability in similarity preservation, resulting in less desirable performance. To address this issue, we propose an asymmetric multi-valued hashing method supported by two different non-binary embeddings. (1) A real-valued embed ding is used for representing the newly-coming query. (2) A multi-integer-embedding is employed for compressing the whole database, which is modeled by binary sparse representation with fixed sparsity. With these two non-binary embeddings, the similarities between data points can be preserved precisely. To perform meaningful asymmetric similarity computation for efficient semantic search, these embeddings are jointly learnt by preserving the label-based similarity. Technically, this results in a mixed integer programming problem, which is efficiently solved by alternative optimization. Extensive experiments on three multi-label datasets demonstrate that our approach not only outperforms the existing binary hashing methods in search accuracy, but also retains their query and storage efficiency. |
收录类别 | EI |
语种 | 英语 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/15222 |
专题 | 多模态人工智能系统全国重点实验室_先进时空数据分析与学习 |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 2.School of Computer and Control Engineering, University of Chinese Academy of Sciences |
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 | Da, Cheng,Xu, Shibiao,Ding, Kun,et al. AMVH: Asymmetric Multi-Valued Hashing[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 | ||
Da_AMVH_Asymmetric_M(678KB) | 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