arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.13258cs.CV
收藏

EpO-Net:利用密集轨迹上的几何约束实现运动显著性

EpO-Net: Exploiting Geometric Constraints on Dense Trajectories for Motion Saliency

Information Technology University(信息技术大学);KeepTruckin, Inc(KeepTruckin公司);Australian National University(澳大利亚国立大学)

Muhammad Faisal, Ijaz Akhter, Mohsen Ali, Richard Hartley

摘要:

现有的显著运动分割方法不能明确地学习几何线索,并且经常对显著的静态对象进行错误检测。我们利用多视图的几何约束,以避免这样的缺点。为了处理像海洋这样的非刚性背景,我们还提出了一种基于运动和外观特征的鲁棒融合机制。我们发现密集的轨迹,覆盖视频中的每个像素,并提出基于几何的核线距离来区分背景和前景区域。轨迹核线距离是数据无关的,并且可以很容易地计算出图像之间的一些特征的对应关系。我们表明,通过结合极线距离与光流,一个强大的运动网络可以学习。使网络能够利用这两个功能,我们提出了一个简单的机制,我们称之为输入丢弃。与仅运动网络相比,我们在DAVIS-2016数据集上的平均IoU得分比之前的最先进网络高出5.2%。通过使用输入丢弃机制将我们的运动网络与外观网络进行鲁棒融合,我们在DAVIS-2016,2017和Segtrackv 2数据集上的表现也优于以前的方法。

英文摘要:

The existing approaches for salient motion segmentation are unable to explicitly learn geometric cues and often give false detections on prominent static objects. We exploit multiview geometric constraints to avoid such shortcomings. To handle the nonrigid background like a sea, we also propose a robust fusion mechanism between motion and appearance-based features. We find dense trajectories, covering every pixel in the video, and propose trajectory-based epipolar distances to distinguish between background and foreground regions. Trajectory epipolar distances are data-independent and can be readily computed given a few features' correspondences between the images. We show that by combining epipolar distances with optical flow, a powerful motion network can be learned. Enabling the network to leverage both of these features, we propose a simple mechanism, we call input-dropout. Comparing the motion-only networks, we outperform the previous state of the art on DAVIS-2016 dataset by 5.2% in the mean IoU score. By robustly fusing our motion network with an appearance network using the input-dropout mechanism, we also outperform the previous methods on DAVIS-2016, 2017 and Segtrackv2 dataset.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑