1 / 8

“GrabCut”

“GrabCut”. Project Progress and Future Plans By: Matthew Marsh. What is “GrabCut”. “GrabCut” is an image segmentation technique My aim is to implement “GrabCut” as a plugin for Gimp Code written in C++ and C. Image Segmentation. Process of Splitting images into parts

rey
Télécharger la présentation

“GrabCut”

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. “GrabCut” Project Progress and Future Plans By: Matthew Marsh

  2. What is “GrabCut” • “GrabCut” is an image segmentation technique • My aim is to implement “GrabCut” as a plugin for Gimp • Code written in C++ and C

  3. Image Segmentation • Process of Splitting images into parts • “GrabCut” uses region and boundary information in the image to perform segemtation • Strong gradients indicate boundaries • Areas of colour coherence indicate regions

  4. How “GrabCut” Works • User Needs to help the algorithm by specifying foreground and background regions • Builds up a Graph where nodes represent pixels • Uses a Max-Flow algorithm to segment the graph, and the represented image

  5. Graph cut

  6. How My Plugin Works • Allows the user to paint on foreground and background regions. • Histrograms created from these “clues” from the user, and then uses the max flow algorithm to segment the image. • At the moment only using Greyscale images. • Use GDK to do all graphical interface

  7. Future Plans • So far only have a greyscale implementation • intend to implement colour • Can’t use histograms, have to use a Gaussian Mixture Model • If I have sufficent time implement a border matting feature to deal with hair and other fine textures.

  8. Guassian Mixture Models • I:\MixtureEM\JavaApp.html

More Related