arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.00375cs.CV
收藏

基于扰动的视频理解网络的可视化解释

Towards Visually Explaining Video Understanding Networks with Perturbation

Institute of Industrial Science, The University of Tokyo(东京大学生产技术研究所);National Institute of Advanced Industrial Science and Technology(产业技术综合研究所);ETH Zürich(苏黎世联邦理工学院)

Zhenqiang Li, Weimin Wang, Zuoyue Li, Yifei Huang, Yoichi Sato

摘要:

“让黑盒模型变得可解释”是伴随着深度学习网络发展的一个重要问题。对于以视觉信息作为输入的网络,一个基本但具有挑战性的解释方法是识别和可视化主导网络预测的输入像素/区域。然而,大多数现有的作品集中在解释网络采取单一的图像作为输入,并没有考虑存在于视频的时间关系。提供一种易于使用的视觉解释方法,适用于视频理解网络的多样化结构仍然是一个开放的挑战。在本文中,我们研究了一个通用的扰动为基础的方法,直观地解释视频理解网络。此外,我们提出了一种新的损失函数,以提高其结果在空间和时间维度上的平滑性的限制的方法。该方法使得不同网络结构之间的解释结果的比较成为可能,并且还可以避免对视频输入产生病态的对抗性解释。实验对比结果验证了该方法的有效性。

英文摘要:

''Making black box models explainable'' is a vital problem that accompanies the development of deep learning networks. For networks taking visual information as input, one basic but challenging explanation method is to identify and visualize the input pixels/regions that dominate the network's prediction. However, most existing works focus on explaining networks taking a single image as input and do not consider the temporal relationship that exists in videos. Providing an easy-to-use visual explanation method that is applicable to diversified structures of video understanding networks still remains an open challenge. In this paper, we investigate a generic perturbation-based method for visually explaining video understanding networks. Besides, we propose a novel loss function to enhance the method by constraining the smoothness of its results in both spatial and temporal dimensions. The method enables the comparison of explanation results between different network structures to become possible and can also avoid generating the pathological adversarial explanations for video inputs. Experimental comparison results verified the effectiveness of our method.

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。

↑