中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 紫东太初大模型研究中心
  5. 图像与视频分析团队
Fine-Grained Human-Centric Tracklet Segmentation with Single Frame Supervision

文献类型：期刊论文

作者Liu, Si4; Ren, Guanghui5; Sun, Yao5; Wang, Jinqiao1; Wang, Changhu3; Li, Bo4; Yan, Shuicheng2
刊名IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
出版日期2022-02-01
卷号44期号:2页码:610-621
关键词Labeling Object segmentation Image segmentation Task analysis Semantics Training Face Video object segmentation human-centric fine-grained optical flow estimation
ISSN号0162-8828
DOI10.1109/TPAMI.2019.2911936
通讯作者Wang, Jinqiao(jqwang@nlpr.ia.ac.cn)
英文摘要In this paper, we target at the Fine-grAined human-Centric Tracklet Segmentation (FACTS) problem, where 12 human parts, e.g., face, pants, left-leg, are segmented. To reduce the heavy and tedious labeling efforts, FACTS requires only one labeled frame per video during training. The small size of human parts and the labeling scarcity makes FACTS very challenging. Considering adjacent frames of videos are continuous and human usually do not change clothes in a short time, we explicitly consider the pixel-level and frame-level context in the proposed Temporal Context segmentation Network (TCNet). On the one hand, optical flow is on-line calculated to propagate the pixel-level segmentation results to neighboring frames. On the other hand, frame-level classification likelihood vectors are also propagated to nearby frames. By fully exploiting the pixel-level and frame-level context, TCNet indirectly uses the large amount of unlabeled frames during training and produces smooth segmentation results during inference. Experimental results on four video datasets show the superiority of TCNet over the state-of-the-arts. The newly annotated datasets can be downloaded via http://liusi-group.com/projects/FACTS for the further studies.
资助项目National Key R&D Program of China[2016YFC0801003] ; Natural Science Foundation of China[U1536203] ; Natural Science Foundation of China[61572493] ; Natural Science Foundation of China[61876177] ; Natural Science Foundation of China[61772527] ; Natural Science Foundation of China[61806200]
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:000740006100006
出版者IEEE COMPUTER SOC
资助机构National Key R&D Program of China ; Natural Science Foundation of China
源URL[http://ir.ia.ac.cn/handle/173211/47190]  
专题自动化研究所_模式识别国家重点实验室_图像与视频分析团队
通讯作者Wang, Jinqiao
作者单位1.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
2.Qihoo 360 AI Inst, Beijing, Peoples R China
3.ByteDance AI Lab, Beijing, Peoples R China
4.Beihang Univ, Sch Comp Sci & Engn, Beijing 100191, Peoples R China
5.Chinese Acad Sci, Inst Informat Engn, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714		 Liu, Si,Ren, Guanghui,Sun, Yao,et al. Fine-Grained Human-Centric Tracklet Segmentation with Single Frame Supervision[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2022,44(2):610-621.
APA Liu, Si.,Ren, Guanghui.,Sun, Yao.,Wang, Jinqiao.,Wang, Changhu.,...&Yan, Shuicheng.(2022).Fine-Grained Human-Centric Tracklet Segmentation with Single Frame Supervision.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,44(2),610-621.
MLA Liu, Si,et al."Fine-Grained Human-Centric Tracklet Segmentation with Single Frame Supervision".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 44.2(2022):610-621.

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。