Global Perception Feedback Convolutional Neural Networks | |
Fu, Chaoyou; Wu, Xiang; Dong, Jing; He, Ran | |
2017 | |
会议名称 | Advances in Image and Graphics Technologies |
会议日期 | 2017 |
会议地点 | Beijing, China |
摘要 | Top-down feedback mechanism is an important module of visual attention for weakly supervised learning. Previous top-down feedback convolutional neural networks often perform local perception during feedback. Inspired by the fact that the visual system is sensitive to global topological properties [1], we propose a global perception feedback convolutional neural network that considers the global structure of visual response during feedback inference. The global perception eliminates “Visual illusions” that are produced in the process of visual attention. It is achieved by simply imposing the trace norm on hidden neuron activations. Particularly, when updating the status of hidden neuron activations during gradient backpropagation, we get rid of some minor constituent in the SVD decomposition, which both ensures the global low-rank structure of feedback information and the elimination of local noise. Experimental results on the ImageNet dataset corroborate our claims and demonstrate the effectiveness of our global perception model. |
语种 | 英语 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/21096 |
专题 | 智能感知与计算研究中心 |
通讯作者 | Fu, Chaoyou |
作者单位 | Institution of Automation, Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Fu, Chaoyou,Wu, Xiang,Dong, Jing,et al. Global Perception Feedback Convolutional Neural Networks[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 | ||
IGTA.pdf(1840KB) | 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