Click to Correction: Interactive Bidirectional Dynamic Propagation Video Object Segmentation Network
文献类型： 外文期刊
作者： Yang, Shuting 1 ; Yuan, Xia 2 ; Luo, Sihan 2 ;
作者机构： 1.Ningxia Acad Agr & Forestry Sci, Inst Agr Econ & Informat Technol, Yinchuan 750002, Peoples R China
2.Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Peoples R China
关键词： video object segmentation; interactive segmentation; click-based interactive; bidirectional propagation
期刊名称：SENSORS （ 影响因子：3.5； 五年影响因子：3.7 ）
ISSN：
年卷期： 2024 年 24 卷 19 期
页码：
收录情况： SCI
摘要： High-quality video object segmentation is a challenging visual computing task. Interactive segmentation can improve segmentation results. This paper proposes a multi-round interactive dynamic propagation instance-level video object segmentation network based on click interaction. The network consists of two parts: a user interaction segmentation module and a bidirectional dynamic propagation module. A prior segmentation network was designed in the user interaction segmentation module to better segment objects of different scales that users click on. The dynamic propagation network achieves high-precision video object segmentation through the bidirectional propagation and fusion of segmentation masks obtained from multiple rounds of interaction. Experiments on interactive segmentation datasets and video object segmentation datasets show that our method achieves state-of-the-art segmentation results with fewer click interactions.
- 相关文献
作者其他论文 更多>>
-
Improving Soil Organic Matter Mapping Using Transfer Learning and Satellite-Simulated Samples From Bare Soil Hyperspectral Imagery
作者：Xu, Xibo;Xu, Xibo;Chen, Yunhao;Yang, Shuting
关键词：Satellites; Soil; Soil measurements; Accuracy; Precipitation; Hyperspectral imaging; Data models; Area measurement; Transfer learning; Satellite images; Bare soil hyperspectral imagery; mapping model; satellite-simulated spectral sample; soil organic matter (SOM); transfer learning (TL)
-
Multimodal Prompt Tuning for Hyperspectral and LiDAR Classification
作者：Liu, Zhengyu;Yuan, Xia;Fu, Guanyiman;Zhao, Chunxia;Xiong, Fengchao;Yang, Shuting
关键词：hyperspectral image; light detection and ranging (LiDAR); multimodal fusion; prompt tuning
-
A remote sensing-based strategy for mapping potentially toxic elements of soils: Temporal-spatial-spectral covariates combined with random forest
作者：Xu, Xibo;Wang, Zeqiang;Wang, Zeqiang;Song, Xiaoning;Zhan, Wenjie;Yang, Shuting;Xu, Xibo
关键词：Soil; Pb; Random forest; Temporal-spatial-spectral covariates; Mapping strategy
-
Estimating the Canopy Nitrogen Content in Maize by Using the Transform-Based Dynamic Spectral Indices and Random Forest
作者：Yang, Shuting;Li, Jianbei;Li, Ji;Zhang, Xuejian;Ma, Cong;Liu, Zhengyu;Ren, Mengyan
关键词：maize canopy leaves; nitrogen contents; spectral estimation; transform-based dynamic spectral indices
