Skip to content

A hybrid speed and radial distance feature descriptor using optical flow approach in HAR

Notifications You must be signed in to change notification settings

gao106/Optical_Flow_HAR

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Optical_Flow_HAR

A Hybrid Speed and Radial Distance Feature Descriptor using Optical Flow Approach in HAR

dataset

weizmann dataset: https://www.wisdom.weizmann.ac.il/~vision/SpaceTimeActions.html

dataset structure:

Weizmann

  • Train_datasets
    • walk
    • bend
    • jump
    • ...
  • Test_datasets
    • walk
    • bend
    • jump
    • ...

The resuslt are updated by extract 72*2(speed)+24(Distance) on Weizmann datasets.

RESULTS TABLE (UPDATED)03/08/2022:

Average accuracy : 94.7%

Method Human action classification
Proposed Method WALK 100%
BEND 86.90%
JACK 96%
PJUMP 100%
RUN 91%
SIDE 100%
WAVE1 92%
WAVE2 92%
walk.mp4

About

A hybrid speed and radial distance feature descriptor using optical flow approach in HAR

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 27.6%
  • Java 18.5%
  • HTML 15.2%
  • C 13.2%
  • M4 11.6%
  • Python 11.5%
  • Other 2.4%