CV Chapter 3 Segmentation
Questions about the lecture 'Computer Vision' of the RWTH Aachen Chapter 3 Segmentation
Questions about the lecture 'Computer Vision' of the RWTH Aachen Chapter 3 Segmentation
Set of flashcards Details
| Flashcards | 53 |
|---|---|
| Language | English |
| Category | Computer Science |
| Level | University |
| Created / Updated | 04.02.2017 / 19.02.2017 |
| Weblink |
https://card2brain.ch/box/20170204_cv_chapter_3_segmentation
|
| Embed |
<iframe src="https://card2brain.ch/box/20170204_cv_chapter_3_segmentation/embed" width="780" height="150" scrolling="no" frameborder="0"></iframe>
|
What holds for the mincut and maxflow problems?
Theorem is that max-flow equals min-cut
What are the time-complexity improvements in min-cut/max-flow history? [3]
1. 1951 Dantzig with O(n²mU)
2. 1997 Goldberg and Rao with O(x*log(n²/m)logU) with x is m^3/2 or n^2/3*m
3. n is #nodes m is #edges and U is maximum edge weight
What is the definition of the maxflow algorithm? [4]
1. Find source→sink path with positive capacity
2. Push maximum possible flow through path
3. Repeat until all paths are full
4. All full edges belong to cut
What are the computer vision applications of maxflow? [2]
1. For cv with grid graphs with low connectivity
2. Dual search tree augmenting path algorithm (Boykov and Kolmogorov 2004)
What is one application using graph cut?
2001 Boykov and Jolly with GrabCut
What are the characteristics of GrabCut? [2]
1. Interactive image segmentation // Fore- and background labeled with brush or bounding box
2. In newest version of MS Office
What are the characteristics of the objects of GrabCut? [2]
1. Object is coherent set of pixels
2. Psi(x,y)=gamma* Sum_mn delta(xn!=xm)*e(-beta*|ym-yn|²) //How to choose gamma?
What are the advantages of graph cut? [4]
1. Powerful based on MRF
2. Applicable for wide range of problems
3. Efficient for cv problems
4. Becoming standard
What are the disadvantages of graph cut? [2]
1. Limited class of models
2. Only approximative for multi-label cases
What is the problem segmentation of non-binary cases? [2]
1. Limitation to binary energies is often nuisance
2. Problem is NP-hard with \geq 3 labels
What are possible solutions for segmentation of non-binary cases?
Approximation algorithms for multi-label cases which are no guarantee for optimum
What are possible approximation algorithms for multi-label cases? [2]
1. alpha-Expansion and 2. alphaBeta-Swap
What is the definition of the alpha-Expansion algorithm? [3]
1. Start with any initial solution
2. For each label compute optima s-t graph cut with energy decrease
3. Stop if no expansion would decrease energy
