[译]Deep Learning for Videos: A 2018 Guide to Action Recognition

原文地址:Deep Learning for Videos: A 2018 Guide to Action Recognition

这是一篇18年的综述性博客,对于视频分类领域的发展有一个较详细的说明

Medical images like MRIs, CTs (3D images) are very similar to videos - both of them encode 2D spatial information over a 3rd dimension. Much like diagnosing abnormalities from 3D images, action recognition from videos would require capturing context from entire video rather than just capturing information from each frame.

像核磁共振成像、计算机断层扫描(3D图像)这样的医学图像非常类似于视频 - 它们都在三维空间编码2D空间信息。就像从3D图像中诊断异常一样,从视频中进行动作识别需要从整个视频中捕捉上下文,而不仅仅是从每一帧中捕捉信息