Knowledge Commons of Institute of Automation,CAS
SparseMask: Differentiable Connectivity Learning for Dense Image Prediction | |
Wu, Huikai1,2; Zhang, Junge1,2; Huang, Kaiqi1,2 | |
2019-10 | |
会议名称 | IEEE International Conference on Computer Vision |
页码 | 6767-6776 |
会议日期 | 27 Oct.-2 Nov. 2019 |
会议地点 | Seoul, Korea (South) |
摘要 | In this paper, we aim at automatically searching an efficient network architecture for dense image prediction. Particularly, we follow the encoder-decoder style and focus on designing a connectivity structure for the decoder. To achieve that, we design a densely connected network with learnable connections, named Fully Dense Network, which contains a large set of possible final connectivity structures. We then employ gradient descent to search the optimal connectivity from the dense connections. The search process is guided by a novel loss function, which pushes the weight of each connection to be binary and the connections to be sparse. The discovered connectivity achieves competitive results on two segmentation datasets, while runs more than three times faster and requires less than half parameters compared to the state-of-the-art methods. An extensive experiment shows that the discovered connectivity is compatible with various backbones and generalizes well to other dense image prediction tasks. |
七大方向——子方向分类 | 图像视频处理与分析 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/38529 |
专题 | 复杂系统认知与决策实验室_智能系统与工程 |
通讯作者 | Huang, Kaiqi |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences |
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 | Wu, Huikai,Zhang, Junge,Huang, Kaiqi. SparseMask: Differentiable Connectivity Learning for Dense Image Prediction[C],2019:6767-6776. |
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 | ||
代表性论文3-武慧凯.pdf(402KB) | 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