Skip to content

Light and Fast Face Detector (LFFD) implementation

Notifications You must be signed in to change notification settings

VectXmy/LFFD.Pytorch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

LFFD: A Light and Fast Face Detector for Edge Devices

paper link

Requirements:

  • pytorch>=1.2
  • opencv-python
  • torchvision>=0.4

Label Assignment

Boxes are in different scale. Anchor points with more than one matched gt box are ignored.
green box --> gt box
blue/red point --> positive sample
The positive sample sampling of the fifth branch is shown below.

Different settings from the paper

  • regression target
  • add BN
  • more classes

Test

I have trained the model on a safety helmet dataset without any augmentation. It seems to be working well.

tensorboard

test_result

About

Light and Fast Face Detector (LFFD) implementation

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages