Skip to content

Human action recognition using deep learning and classical machine learning algorithms

License

Notifications You must be signed in to change notification settings

saurabhkar/action-recognition

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 

Repository files navigation

action-recognition

The major final year project.ipynb extract data from HMDB51 dataset and process it using various algorithms.

recognition.py is the unchanged code and no modification has been done on that.

link to the complete dataset can be downloaded from here. CLICK HERE .

  • Some random images were taken from each class ( total number of classes =51) and datafame was made using those images.
  • SIFT algorithm was applied on that dataframe to extract the SIF points (keypoints and descriptors).
  • HOG was applied on the original dataframe to extract HOG features.
  • Both the above algorithms features have been classified using SVM and DNN.
  • SIFT and HOG feature combination has also been implemented and classified using SVM and DNN.
  • CNN was used independently directly on the original dataframe.

About

Human action recognition using deep learning and classical machine learning algorithms

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published