Rethinking the Faster R-CNN Architecture for Temporal Action Localization

版权声明：添加我的微信wlagooble，开启一段不一样的旅程 https://blog.csdn.net/nineship/article/details/86308167

论文：Rethinking the Faster R-CNN Architecture for Temporal Action Localization

CVPR 2018

link: http://cn.arxiv.org/pdf/1804.07667.pdf

摘要：

主要贡献有下面三个

1. we improve receptive field alignment using a multi-scale architecture that can accommodate extreme variation in action durations;

2. we better exploit the temporal context of actions for both proposal generation and action classification by appropriately extending receptive fields

3. we explicitly consider multi-stream feature fusion and demonstrate that fusing motion late is important

介绍：

扫描二维码关注公众号，回复： 5181327 查看本文章

除了识别出动作，还要识别出动作的开始结束时间

 where the task is to not only identify the action class, but also detect the start and end time of each action instance

相关工作：

1.Action Recognition 动作识别

Tremendous progress has recently been made due to the introduction of large datasets and the developments on deep neural networks [37, 30, 43, 49, 7, 14]深度学习的发展。

2.Temporal Action Localization 时间动作定位

Yuan et al. [54] proposed a multi-scale pooling scheme to capture features at multiple resolutions.

many recent approaches adopt a two-stage, proposal-plus-classification framework [6, 36,12, 3, 4, 35, 56]

3 Faster R-CNN

数据集：THUMOS’14 

未完待续