Skip to content

Using opencv3 and cuda to extract optical flow image frame by frame

Notifications You must be signed in to change notification settings

TyroneLi/Dense-Flow-GPU

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Dense-Flow-GPU

1.usage:

use opencv 3 to extract optical flow image frame-by-frame in videos.Such as flow_x and flow_y. Some times use these code to get optical flow frame by frame to train models. Enjoy it.

2.command use:

./denseFlow_gpu --vidFile=<input_video_path> --xFlowFile=<optical_flow_save_path>/flow_x --yFlowFile=<optical_flow_save_path>/flow_y --imgFile=<extracted_frames_save_path>/ --bound=20 --type=1 --device_id=7 --step=1

--bound:clip mv value inside [-20, 20], --step:extract informations every n steps.

3.update use:

  • update DensePyrLKOpticalFlow algorithm 💪
  • update FarnebackOpticalFlow algorithm 💪
  • update OpticalFlowDual_TVL1 algorithm 💪
  • update BroxOpticalFlow algorithm 💪

About

Using opencv3 and cuda to extract optical flow image frame by frame

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published