Skip to content
/ ConvLSTM-action-recognition Public

Simple video-based action recognition pipeline using ConvLSTM

3 stars 1 fork Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

farhantandia/ConvLSTM-action-recognition

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits

.vscode

.vscode

 
 

youtube data/zoomed_data_raw

youtube data/zoomed_data_raw

 
 

.gitattributes

.gitattributes

 
 

.gitignore

.gitignore

 
 

README.md

README.md

 
 

action-recognition-swim.py

action-recognition-swim.py

 
 

combine.py

combine.py

 
 

debug.log

debug.log

 
 

ffmpeg-split.py

ffmpeg-split.py

 
 

image.jpg

image.jpg

 
 

imgtovid.py

imgtovid.py

 
 

network.jpg

network.jpg

 
 

results.txt

results.txt

 
 

split.py

split.py

 
 

Repository files navigation

Action recognition using ConvLSTM

Simple video-based action recognition pipeline using ConvLSTM

Dependencies

  • Tensorflow 2.3
  • OpenCV
  • keras-video-generators

Network


1. 3D input images 2. Multiple stacks of ConvLSTM along with batch normalization layer and with dropout 20% 3. Dense layers=> Final NN has 9.8M parameters with accuracy by 94.97%.

About

Simple video-based action recognition pipeline using ConvLSTM

Topics

actionrecognition convlstm-network

Resources

Readme
Activity

Stars

3 stars

Watchers

1 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

No packages published

Languages