Video Recovery via Learning Variation and Consistency of Images
Video Recovery via Learning Variation and Consistency of Images
复制标题
DOI：
10.1609/aaai.v31i1.11241
复制
发表时间：
2017-02
期刊：
影响因子：
--
通讯作者：
Zhouyuan Huo;Shangqian Gao;Weidong (Tom) Cai;Heng Huang
中科院分区：
文献类型：
--
作者：
Zhouyuan Huo;Shangqian Gao;Weidong (Tom) Cai;Heng Huang
矩阵补全算法已被广泛地应用于缺项图像的恢复，并被证明是非常有效的。最近的研究在视频恢复中使用了张量补全模型，假设所有视频帧都是均匀且相关的。而真实的视频是由不同的片段或场景组成的，即异质性。因此，需要一种同时利用视频时空一致性和变化性的视频恢复模型。为了解决这一问题，我们提出了一种新的带有变化和一致性约束的分段跟踪范数（STN-VCC）视频恢复方法。在我们的模型中，使用上限l1范数正则化来学习视频片段中连续帧之间的时空一致性和变化。同时，我们引入了一种新的低秩模型来捕获视频帧中的低秩结构，它比传统的迹范数具有更好的秩最小化近似。提出了一种高效的优化算法，并给出了收敛性证明。我们通过几个视频恢复任务对所提出的方法进行了评估，实验结果表明，我们的新方法始终优于其他相关方法。
Matrix completion algorithms have been popularly used to recover images with missing entries, and they are proved to be very effective. Recent works utilized tensor completion models in video recovery assuming that all video frames are homogeneous and correlated. However, real videos are made up of different episodes or scenes, i.e. heterogeneous. Therefore, a video recovery model which utilizes both video spatiotemporal consistency and variation is necessary. To solve this problem, we propose a new video recovery method Sectional Trace Norm with Variation and Consistency Constraints (STN-VCC). In our model, capped L1-norm regularization is utilized to learn the spatial-temporal consistency and variation between consecutive frames in video clips. Meanwhile, we introduce a new low-rank model to capture the low-rank structure in video frames with a better approximation of rank minimization than traditional trace norm. An efficient optimization algorithm is proposed, and we also provide a proof of convergence in the paper. We evaluate the proposed method via several video recovery tasks and experiment results show that our new method consistently outperforms other related approaches.