Robust Video Object Segmentation via Propagating Seams and Matching Superpixels
Robust Video Object Segmentation via Propagating Seams and Matching Superpixels
复制标题
通过传播接缝和匹配超像素进行稳健的视频对象分割
DOI：
10.1109/access.2020.2981140
复制
发表时间：
2020
期刊：
影响因子：
3.9
通讯作者：
Liu Caixing
中科院分区：
文献类型：
--
作者：
Liang Yun;Zhang Yuqing;Wu Yihan;Tu Shuqin;Liu Caixing
视频对象分割的目的是将前景对象从背景中分离出来，但由于存在变形、遮挡和运动模糊等问题，目前还没有很好的解决。提出了一种基于拼接缝传播和超像素匹配的鲁棒视频对象分割方法。首先，我们预测的初始对象的轮廓基于像素级的目标标签计算补丁缝传播和粗糙集。通过一个面片接缝，将当前面片映射到上一帧中与当前面片最相似的面片上，并根据映射后的面片标签得到当前面片的标签。其次，我们利用超像素作为中间级别的线索，以优化预测的对象轮廓。提供基于三个亮度通道的双向距离以匹配相邻帧之间的超像素。利用匹配结果的边界和初始化的目标轮廓，构造多个候选目标轮廓。然后定义一个基于多特征的能量函数来度量候选轮廓，能量最小的轮廓即为当前帧的最终分割结果。最后，通过传播补丁接缝和匹配超像素，我们计算视频对象分割结果逐帧。SegTrack-v2数据的14个视频用于评估我们的方法。定量和定性的评价表明，我们的方法比大多数现有的方法，特别是在处理遮挡，变形和运动模糊。
Video object segmentation aims at separating foreground object from background, and it is far from well solved for different challenges such as deformation, occlusion and motion blurs. This paper proposes a robust video object segmentation method by propagating patch seams and matching superpixels. First, we predict the initial object contour based on pixel-level target labels calculated by patch seam propagation and rough sets. By a patch seam, we map a current patch to its most similar patch from last frame and obtain its labels based on the labels of mapped patch. Second, we utilize superpixels as middle level cues to optimize predicted object contour. The bidirectional distance based on three brightness channels is provided to match superpixels between adjacent frames. Using the boundaries of matched results and initialized object contour, many candidates of object contours are constructed. Third, we define an energy function based on multi-features to measure contour candidates, and the contour with minimum energy is the final segmented result of current frame. Finally, by propagating patch seams and matching superpixels, we compute video object segmentation results frame by frame. Fourteen videos of SegTrack-v2 data are used to evaluate our method. The quantitative and qualitative evaluations show that our method performs better than most present methods especially in dealing with occlusion, deformation and motion blurs.
登录
查看更多内容
DOI：
10.1109/cvpr.2015.7298835
发表时间：
2015-06
期刊：
2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
影响因子：
--
作者：
Longyin Wen;Dawei Du;Zhen Lei;S. Li;Ming-Hsuan Yang
通讯作者：
Longyin Wen;Dawei Du;Zhen Lei;S. Li;Ming-Hsuan Yang
影响因子：
3.9
作者：
Jiangyun Li;Yikai Zhao;J. Fu;Jiajia Wu;Jing Liu
通讯作者：
Jiangyun Li;Yikai Zhao;J. Fu;Jiajia Wu;Jing Liu
影响因子：
10.6
作者：
Atilla Sit;Daisuke Kihara
通讯作者：
Atilla Sit;Daisuke Kihara
DOI：
10.1109/icpr.2018.8545723
发表时间：
2018-04
期刊：
2018 24th International Conference on Pattern Recognition (ICPR)
影响因子：
--
作者：
Hajar Sadeghi Sokeh;Vasileios Argyriou;D. Monekosso;Paolo Remagnino
通讯作者：
Hajar Sadeghi Sokeh;Vasileios Argyriou;D. Monekosso;Paolo Remagnino
DOI：
10.1109/cvpr.2013.87
发表时间：
2013-06
期刊：
2013 IEEE Conference on Computer Vision and Pattern Recognition
影响因子：
--
作者：
Dong Zhang;O. Javed;M. Shah
通讯作者：
Dong Zhang;O. Javed;M. Shah