Knowledge Commons of Institute of Automation,CAS
Two-Stream Gated Fusion ConvNets for Action Recognition | |
Zhu, Jiagang1,2![]() ![]() ![]() | |
2018-11 | |
会议名称 | 2018 24th International Conference on Pattern Recognition (ICPR) |
会议日期 | 20-24 Aug. 2018 |
会议地点 | Beijing, China |
摘要 | The two-stream ConvNets in action recognition always fuse the two streams' predictions by the weighted averaging scheme. This fusion way with fixed weights lacks of pertinence to different action videos and always needs trial and error on the validation set. In order to enhance the adaptability of two-stream ConvNets, an end-to-end trainable gated fusion method, namely gating ConvNet, is proposed in this paper based on the MoE (Mixture of Experts) theory. The gating ConvNet takes the combination of convolutional layers of the spatial and temporal nets as input and outputs two fusion weights. To reduce the over-fitting of gating ConvNet caused by the redundancy of parameters, a new multi-task learning method is designed, which jointly learns the gating fusion weights for the two streams and learns the gating ConvNet for action classification. With the proposed gated fusion method and multi-task learning approach, competitive performance is achieved on the video action dataset UCF101. |
收录类别 | EI |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/39108 |
专题 | 中科院工业视觉智能装备工程实验室_精密感知与控制 |
通讯作者 | Zou, Wei |
作者单位 | 1.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China |
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 | Zhu, Jiagang,Zou, Wei,Zhu, Zheng. Two-Stream Gated Fusion ConvNets for Action Recognition[C],2018. |
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 | ||
Two-Stream Gated Fus(943KB) | 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