Two-Stream Convolutional Networksfor Action Recognition in Videos | 大海

[zjykzj]

https://blog.zhujian.life/posts/fa4c1da.html

原文地址：Learning Spatiotemporal Features with 3D Convolutional Networks