interactive image segmentation sara vicente 1 ([email protected])

1
Interactive image segmentation Sara Vicente 1 ([email protected]) Supervised by Vladimir Kolmogorov 1 and Carsten Rother 2 1 University College London, 2 Microsoft Research Cambridge The aim of interactive image segmentation is to extract an object from an image by segmenting the image in two regions: background and foreground. To minimize the problems of fully automatic segmentation, a user imposes some hard constraints: a lasso or rectangle around the object or the specification of regions that have to be part of background or foreground. GrabCut overview T F T U T B Inpu t Assign to each pixel a label 0 – background, 1 – foreground dividing the image in two regions Colour agreement : colour of the pixel should agree with the colour model of the label assigned to it (colour models are computed for background and foreground) Regional coherence : neighbour pixels should be assigned the same label, especially if the colour of both is similar. Different weights can be given to the two components of the model producing very distinct results. Computes segmentation using a standard minimum cut algorithm Updates in each iteration the colour model for background and foreground based on last iteration First iteration Last iteration Extreme settings: exaggerated colour agreement weight User Input Trimap: T F – foreground T U – unknown region T B – background Goal Model Constraints Iterative algorithm For some images, GrabCut algorithm has a shrinking effect, cutting elongated structures. It was proven in [1] that it is possible to integrate the optimization of flux in the GrabCut framework. This integration should prevent this shrinking effect to happen. The choice of the vector field for which we intend to optimize the flux should be done carefully in order to achieved the desirable results. Development and test of new vector fields that can be used for the flux optimization Learn parameters of the model: weights of the different components (agreement with data, regional coherence and flux) Evaluation of the new model using a more complete database of images Future work Reference s: [1] Vladimir Kolmogorov and Yuri Boykov. What metrics can be approximated by geocuts, or global optimization of length/area and flux. In ICCV ’05, 2005. [2] C. Rother, V. Komogorov, and A. Blake, “GrabCut” - Interactive foreground extraction using iterated graph cuts. In ACM Transactions on Graphics (SIGGRAPH'04), 2004 GrabCut “shrinking” effect Results with flux Improving GrabCut: introducing flux Extreme settings: exaggerated regional coherence weight

Upload: jaser

Post on 10-Feb-2016

48 views

Category:

Documents


0 download

DESCRIPTION

T B. T U. T F. Interactive image segmentation Sara Vicente 1 ([email protected]) Supervised by Vladimir Kolmogorov 1 and Carsten Rother 2 1 University College London, 2 Microsoft Research Cambridge. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Interactive image segmentation Sara Vicente  1 (s.vicente@adastral.ucl.ac.uk)

Interactive image segmentationSara Vicente 1 ([email protected])

Supervised by Vladimir Kolmogorov 1 and Carsten Rother 2

1 University College London, 2 Microsoft Research Cambridge

The aim of interactive image segmentation is to extract an object from an image by segmenting the image in two regions: background and foreground.

To minimize the problems of fully automatic segmentation, a user imposes some hard constraints: a lasso or rectangle around the object or the specification of regions that have to be part of background or foreground.

GrabCut overview

TF

TU

TB

InputAssign to each pixel a label

0 – background, 1 – foreground

dividing the image in two regions

Colour agreement: colour of the pixel should agree with the colour model of the label assigned to it (colour models are computed for background and foreground)

Regional coherence : neighbour pixels should be assigned the same label, especially if the colour of both is similar.

Different weights can be given to the two components of the model producing very distinct results.

Computes segmentation using a standard minimum cut algorithm

Updates in each iteration the colour model for background and foreground based on last iteration

First iteration

Last iterationExtreme settings: exaggerated colour agreement weight

User Input Trimap: TF – foregroundTU – unknown regionTB – background

Goal Model Constraints Iterative algorithm

For some images, GrabCut algorithm has a shrinking effect, cutting elongated structures.

It was proven in [1] that it is possible to integrate the optimization of flux in the GrabCut framework. This integration should prevent this shrinking effect to happen.

The choice of the vector field for which we intend to optimize the flux should be done carefully in order to achieved the desirable results.

Development and test of new vector fields that can be used for the flux optimization

Learn parameters of the model: weights of the different components (agreement with data, regional coherence and flux)

Evaluation of the new model using a more complete database of images

Future work References:[1] Vladimir Kolmogorov and Yuri Boykov. What metrics can be approximated by

geocuts, or global optimization of length/area and flux. In ICCV ’05, 2005.

[2] C. Rother, V. Komogorov, and A. Blake, “GrabCut” - Interactive foreground extraction using iterated graph cuts. In ACM Transactions on Graphics (SIGGRAPH'04), 2004

GrabCut “shrinking” effect

Results with flux

Improving GrabCut: introducing flux

Extreme settings: exaggerated regional coherence weight