Skip to content

Given a panoramic sequence of camera images across five cameras and over time, produce a set of panoptic segmentation labels for each pixel, where the instance labels are consistent across all images in the sequence.

License

Notifications You must be signed in to change notification settings

aryaman-patel/DETR-PS

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Panoptic-Segmentation

Implemented an end-to-end transformer based model (DETR) for the task of Panoptic segmentation.

Given Sequence:

Image_seq

Current output of the model:

Image_seq

TODO:

  • Extend this model for instance matching over sequences of images, i.e. Video Panoptic segmentation, perhaps using something from ViP DeepLab?

About

Given a panoramic sequence of camera images across five cameras and over time, produce a set of panoptic segmentation labels for each pixel, where the instance labels are consistent across all images in the sequence.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 94.9%
  • Python 5.1%