arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1904.09288cs.CV
收藏

STEP:用于视频动作检测的时空渐进学习

STEP: Spatio-Temporal Progressive Learning for Video Action Detection

University of Maryland, College Park(马里兰大学帕克分校);NVIDIA(英伟达);University of California, Davis(加利福尼亚大学戴维斯分校)

Xitong Yang, Xiaodong Yang, Ming-Yu Liu, Fanyi Xiao, Larry Davis, Jan Kautz

摘要:

在本文中,我们提出了空间-时间渐进(STEP)动作检测器-一个渐进的学习框架,用于视频中的时空动作检测。从一些粗略的建议长方体开始,我们的方法通过几个步骤逐步完善行动的建议。这样,高质量的提案(即,坚持动作运动)可以在稍后的步骤中通过利用来自先前步骤的回归输出来逐渐获得。在每一步中,我们自适应地扩展时间的建议,以纳入更多相关的时间背景。与以前在一次运行中执行动作检测的工作相比,我们的渐进式学习框架能够自然地处理动作管内的空间位移,因此为时空建模提供了更有效的方法。我们广泛地评估了我们的方法在UCF 101和AVA上,并展示了优越的检测结果。值得注意的是,我们在两个数据集上实现了75.0%和18.6%的mAP,采用3个渐进步骤,分别仅使用11个和34个初始提案。

英文摘要:

In this paper, we propose Spatio-TEmporal Progressive (STEP) action detector---a progressive learning framework for spatio-temporal action detection in videos. Starting from a handful of coarse-scale proposal cuboids, our approach progressively refines the proposals towards actions over a few steps. In this way, high-quality proposals (i.e., adhere to action movements) can be gradually obtained at later steps by leveraging the regression outputs from previous steps. At each step, we adaptively extend the proposals in time to incorporate more related temporal context. Compared to the prior work that performs action detection in one run, our progressive learning framework is able to naturally handle the spatial displacement within action tubes and therefore provides a more effective way for spatio-temporal modeling. We extensively evaluate our approach on UCF101 and AVA, and demonstrate superior detection results. Remarkably, we achieve mAP of 75.0% and 18.6% on the two datasets with 3 progressive steps and using respectively only 11 and 34 initial proposals.

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。

↑