Knowledge Commons of Institute of Automation,CAS
Fixed-point Factorized Networks | |
Wang, Peisong1,2; Cheng, Jian1,2,3 | |
2017 | |
会议名称 | IEEE Conference on Computer Vision and Pattern Recognition |
页码 | 4012-4020 |
会议日期 | 2017.7.21-7.26 |
会议地点 | Hawaii,USA |
摘要 |
In recent years, Deep Neural Networks (DNN) based methods have achieved remarkable performance in a wide range of tasks and have been among the most powerful and widely used techniques in computer vision. However, DNN-based methods are both computational-intensive and resource-consuming, which hinders the application of these methods on embedded systems like smart phones. To alleviate this problem, we introduce a novel Fixed-point Factorized Networks (FFN) for pretrained models to reduce the computational complexity as well as the storage requirement of networks. The resulting networks have only weights of -1, 0 and 1, which significantly eliminates the most resource-consuming multiply-accumulate operations (MACs). Extensive experiments on large-scale ImageNet classification task show the proposed FFN only requires one-thousandth of multiply operations with comparable accuracy. |
关键词 | Convolutional Neural Networks Ternary Quantization Network Acceleration Network Compression |
收录类别 | EI |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/20114 |
专题 | 紫东太初大模型研究中心_图像与视频分析 |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences 3.Center for Excellence in Brain Science and Intelligence Technology |
第一作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Wang, Peisong,Cheng, Jian. Fixed-point Factorized Networks[C],2017:4012-4020. |
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 | ||
王培松_定点分解网络.pdf(279KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
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 |
[Wang, Peisong]的文章 |
[Cheng, Jian]的文章 |
百度学术 |
百度学术中相似的文章 |
[Wang, Peisong]的文章 |
[Cheng, Jian]的文章 |
必应学术 |
必应学术中相似的文章 |
[Wang, Peisong]的文章 |
[Cheng, Jian]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