Skip to content

Code for ICCV'17 "A Multilayer-Based Framework for Online Background Subtraction with Freely Moving Cameras"

License

Notifications You must be signed in to change notification settings

EthanZhu90/MultilayerBSMC_ICCV17

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MultilayerBSMC

This is an implementation of "A Multilayer-Based Framework for Online Background Subtraction with Freely Moving Cameras" ICCV'17 with Matlab. You are always welcome to use it as a to-compare method with an uncommercial goal.

Data

Test Dataset: cars1~8, people1, people2 from Berkeley Motion Segmentation Dataset (BMS-26)

Put the testing data to the folder './Data/moseg_dataset/'
To prepare the data, you need to convert the images to 'ppm' format, and prepare addtional 'XXX.bmf' file.
The example is shown in 'cars1'.

Optical Flow

The optical flow I am using is pretty slow. To speed up, you are suggested to use faster deep-learning-based method. OR you can directly use the trajectory-generating methods.
I will add it to my TODO list.

Code

To run the code, check "runHopkins.m".

Result from paper

alt tag

About

Code for ICCV'17 "A Multilayer-Based Framework for Online Background Subtraction with Freely Moving Cameras"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages