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 1810.01455cs.CV
收藏

动作识别的表示流程

Representation Flow for Action Recognition

Indiana University(印第安纳大学)

AJ Piergiovanni, Michael S. Ryoo

摘要:

在本文中,我们提出了一个受光流算法启发的卷积层来学习运动表示。我们的表示流层是一个完全可微的层,旨在捕获卷积神经网络内任何表示通道的“流”,以进行动作识别。其用于迭代流优化的参数与其他CNN模型参数一起以端到端的方式学习,从而最大化动作识别性能。此外,我们新引入的概念,通过堆叠多个表示流层学习“流的流”表示。我们进行了广泛的实验评估,证实了它的优势,比以前的识别模型,使用传统的光流在计算速度和性能。代码/型号可在这里:https://piergiaj.github.io/rep-flow-site/

英文摘要:

In this paper, we propose a convolutional layer inspired by optical flow algorithms to learn motion representations. Our representation flow layer is a fully-differentiable layer designed to capture the `flow' of any representation channel within a convolutional neural network for action recognition. Its parameters for iterative flow optimization are learned in an end-to-end fashion together with the other CNN model parameters, maximizing the action recognition performance. Furthermore, we newly introduce the concept of learning `flow of flow' representations by stacking multiple representation flow layers. We conducted extensive experimental evaluations, confirming its advantages over previous recognition models using traditional optical flows in both computational speed and performance. Code/models available here: https://piergiaj.github.io/rep-flow-site/

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。

↑