Egok360: A 360 Egocentric Kinetic Human Activity Video Dataset
Egok360: A 360 Egocentric Kinetic Human Activity Video Dataset
复制标题
DOI：
10.1109/icip40778.2020.9191256
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Keshav Bhandari;Mario A. DeLaGarza;Ziliang Zong;Hugo Latapie;Yan Yan-Yan
中科院分区：
文献类型：
--
作者：
Keshav Bhandari;Mario A. DeLaGarza;Ziliang Zong;Hugo Latapie;Yan Yan-Yan
近年来，可穿戴传感器的研究受到越来越多的关注，它为360 °视频分析提供了新的研究视角。然而，文献中缺乏360 °数据集，阻碍了该领域的研究。为了弥补这一差距，在本文中，我们提出了一种新的以自我为中心（第一人称）的360°动态人类活动视频数据集（EgoK360）。EgoK360数据集包含具有不同子动作的人类活动的注释，例如，乒乓球运动包括捡球、击球、弹球和发球四个子动作。据我们所知，EgoK360是第一个具有360°环境设置的第一人称活动识别领域的数据集，这将有助于以自我为中心的360 °视频理解。我们提供了实验结果和全面分析的两个流网络的变体360自我中心的活动识别。EgoK360数据集可以从https://egok360.github.io/下载。
Recently, there has been a growing interest in wearable sensors which provides new research perspectives for 360 ° video analysis. However, the lack of 360 ° datasets in literature hinders the research in this field. To bridge this gap, in this paper we propose a novel Egocentric (first-person) 360° Kinetic human activity video dataset (EgoK360). The EgoK360 dataset contains annotations of human activity with different sub-actions, e.g., activity Ping-Pong with four sub-actions which are pickup-ball, hit, bounce-ball and serve. To the best of our knowledge, EgoK360 is the first dataset in the domain of first-person activity recognition with a 360° environmental setup, which will facilitate the egocentric 360 ° video understanding. We provide experimental results and comprehensive analysis of variants of the two-stream network for 360 egocentric activity recognition. The EgoK360 dataset can be downloaded from https://egok360.github.io/.