QuickTopic free message boards logo
Skip to Messages

TOPIC:

Interactive Graph Cuts for Optimal Boundary and Region Segmentation

9
Kristoffer
12-01-2016
05:05 AM ET (US)
GoWin Mobile
  Spam messages 8-7 deleted by QuickTopic between 12-01-2016 05:34 AM and 01-03-2007 02:05 AM
6
Steve Scher
10-21-2004
04:46 PM ET (US)
The histogram for "object" and "background" intensities seems an important part of the finding values for the boundary and region penalites (page 5).

I'm curious whether only a rough estimate is required (dark object on light background, vs light object on dark background), and how sensitive these penalty terms are to choices of seed pixels that are representative of the whole object.
5
Hamed Masnadi-Shirazi
10-21-2004
01:49 PM ET (US)
What is meant by "Isolated Connected Blobs" ?
(fourth page first paragraph)
Edited 10-21-2004 01:53 PM
4
Rasit Topaloglu
10-21-2004
12:02 PM ET (US)
It seems to me that the hard constraints can be overruled, as they are not assigned an infinite cost. But if this is the case, the user assigned bits are not believed to be 100% true. It is interesting that the algorithm doesn't trust the user.

Also, why separating two objects from each other is not an issue but we are only interested in separation of background from objects (I think because it is more difficult)?
Edited 10-21-2004 12:15 PM
3
Louka Dlagnekov
10-21-2004
01:45 AM ET (US)
How would the graph cut work for a 3x3 image that had a horizontal "bar" in the center:

0 0 0
1 1 1
0 0 0

where 0 is background and 1 is object. If you have time tomorrow, could you walk through this example?
2
Robin Hewitt
10-20-2004
07:04 PM ET (US)
Their gestalt example indicates that the behavior of this method can depend heavily on the value for lambda (the region term). Would the user set this value?
1
Stephen Krotosky
10-20-2004
03:40 PM ET (US)
This type of algorithm could be useful for research in situations where people need good segmentation to accomplish other tasks such as object regonition or modeling. The results are obviously better than simple background subtraction or color-based segmentations and could be used to generate "good" data. However, in practice, especially in their video example, this sort of segmentation seems impractical.

Print | RSS Views: 3467 (Unique: 1289 ) / Subscribers: 0 | What's this?