Sparsifying Neural Network Connections for Face Recognition
文献类型：会议论文
| 作者 | Yi Sun; Xiaogang Wang; Xiaoou Tang |
| 出版日期 | 2016 |
| 会议名称 | CVPR2016 |
| 会议地点 | 美国 |
| 英文摘要 | This paper proposes to learn high-performance deep ConvNets with sparse neural connections, referred to as s- parse ConvNets, for face recognition. The sparse ConvNets are learned in an iterative way, each time one additional layer is sparsified and the entire model is re-trained given the initial weights learned in previous iterations. One important finding is that directly training the sparse Con- vNet from scratch failed to find good solutions for face recognition, while using a previously learned denser model to properly initialize a sparser model is critical to continue learning effective features for face recognition. This paper also proposes a new neural correlation-based weight se- lection criterion and empirically verifies its effectiveness in selecting informative connections from previously learned models in each iteration. When taking a moderately sparse structure (26%-76% of weights in the dense model), the proposed sparse ConvNet model significantly improves the face recognition performance of the previous state-of-the- art DeepID2+ models given the same training data, while it keeps the performance of the baseline model with only 12% of the original parameters. |
| 收录类别 | EI |
| 语种 | 英语 |
| 源URL | [http://ir.siat.ac.cn:8080/handle/172644/10022]  |
| 专题 | 深圳先进技术研究院_集成所 |
| 作者单位 | 2016 |
| 推荐引用方式 GB/T 7714 | Yi Sun,Xiaogang Wang,Xiaoou Tang. Sparsifying Neural Network Connections for Face Recognition[C]. 见:CVPR2016. 美国. |
入库方式： OAI收割
来源：深圳先进技术研究院
浏览0
下载0
收藏0
其他版本
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。