ConnNet: A Long-Range Relation-Aware Pixel-Connectivity Network for Salient Segmentation
文献类型：期刊论文
| 作者 | Kampffmeyer, Michael1,2; Dong, Nanqing3,4; Liang, Xiaodan5; Zhang, Yujia6,7 ; Xing, Eric P.8
|
| 刊名 | IEEE TRANSACTIONS ON IMAGE PROCESSING
 |
| 出版日期 | 2019-05-01 |
| 卷号 | 28期号:5页码:2518-2529 |
| 关键词 | Salient segmentation convolutional neural networks salient instance-level segmentation connectivity |
| ISSN号 | 1057-7149 |
| DOI | 10.1109/TIP.2018.2886997 |
| 通讯作者 | Liang, Xiaodan() |
| 英文摘要 | Salient segmentation aims to segment out attention-grabbing regions, a critical yet challenging task and the foundation of many high-level computer vision applications. It requires semantic-aware grouping of pixels into salient regions and benefits from the utilization of global multi-scale contexts to achieve good local reasoning. Previous works often address it as two-class segmentation problems utilizing complicated multi-step procedures, including refinement networks and complex graphical models. We argue that semantic salient segmentation can instead be effectively resolved by reformulating it as a simple yet intuitive pixel-pair-based connectivity prediction task. Following the intuition that salient objects can be naturally grouped via semantic-aware connectivity between neighboring pixels, we propose a pure Connectivity Net (ConnNet). ConnNet predicts the connectivity probabilities of each pixel with its neighboring pixels by leveraging multi-level cascade contexts embedded in the image and long-range pixel relations. We investigate our approach on two tasks, namely, salient object segmentation and salient instance-level segmentation, and illustrate that consistent improvements can be obtained by modeling these tasks as connectivity instead of binary segmentation tasks for a variety of network architectures. We achieve the state-of-the-art performance, outperforming or being comparable to existing approaches while reducing inference time due to our less complex approach. |
| 资助项目 | Norwegian Research Council FRIPRO[239844] |
| WOS研究方向 | Computer Science ; Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:000458850800010 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 资助机构 | Norwegian Research Council FRIPRO |
| 源URL | [http://ir.ia.ac.cn/handle/173211/25032]  |
| 专题 | 自动化研究所_复杂系统管理与控制国家重点实验室_先进机器人控制团队 |
| 通讯作者 | Liang, Xiaodan |
| 作者单位 | 1.Carnegie Mellon Univ, Pittsburgh, PA 15213 USA 2.Univ Tromso Arctic Univ Norway, Machine Learning Grp, N-9019 Tromso, Norway 3.Cornell Univ, Dept Stat Sci Comp & Informat Sci, Ithaca, NY 14850 USA 4.Petuum Inc, Pittsburgh, PA 15222 USA 5.Sun Yat Sen Univ, Sch Intelligent Syst Engn, Guangzhou 510006, Guangdong, Peoples R China 6.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 7.Univ Chinese Acad Sci, Sch Comp & Control Engn, Beijing 100190, Peoples R China 8.Carnegie Mellon Univ, Machine Learning Dept, Pittsburgh, PA 15213 USA |
| 推荐引用方式 GB/T 7714 | Kampffmeyer, Michael,Dong, Nanqing,Liang, Xiaodan,et al. ConnNet: A Long-Range Relation-Aware Pixel-Connectivity Network for Salient Segmentation[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2019,28(5):2518-2529. |
| APA | Kampffmeyer, Michael,Dong, Nanqing,Liang, Xiaodan,Zhang, Yujia,&Xing, Eric P..(2019).ConnNet: A Long-Range Relation-Aware Pixel-Connectivity Network for Salient Segmentation.IEEE TRANSACTIONS ON IMAGE PROCESSING,28(5),2518-2529. |
| MLA | Kampffmeyer, Michael,et al."ConnNet: A Long-Range Relation-Aware Pixel-Connectivity Network for Salient Segmentation".IEEE TRANSACTIONS ON IMAGE PROCESSING 28.5(2019):2518-2529. |
入库方式： OAI收割
来源：自动化研究所
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。