Skip to content

This is the demo of temporal action recognition which we use R3D and SlowFast model. The dataset we use is the UCF-101, you can extract from the UCF-101.zip file.

Notifications You must be signed in to change notification settings

LAKawai/Action-Recognition-by-R3D-and-SlowFast

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Action-Recognition-by-R3D-and-SlowFast

This is the demo of temporal action recognition which we use R3D and SlowFast model.

The dataset we use is the UCF-101, you can get from this link: BaiDuYun Link.

In the train.py, you can select to use R3D or SlowFast model. R3D may need more time to run but show a bit better performance. It is just up to you!

You should save the states of dicts of your models which have finished training into the checkpoint dir, and then, you can use them to have a test. Or you may use our pretrained model, the link of R3D is BaiDuYun Link, and the link of SlowFast is BaiDuYun Link.

About

This is the demo of temporal action recognition which we use R3D and SlowFast model. The dataset we use is the UCF-101, you can extract from the UCF-101.zip file.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages