2/6/2017 1
Segmentation & Grouping
Kristen Grauman UT Austin
Tues Feb 7
Announcements
- A0 on Canvas
- No office hours today
– TA office hours this week as usual
- Guest lecture Thursday by Suyog Jain
– Interactive segmentation
- Check in on pace
Segmentation & Grouping Kristen Grauman UT Austin Tues Feb 7 - - PDF document
2/6/2017 Segmentation & Grouping Kristen Grauman UT Austin Tues Feb 7 Announcements A0 on Canvas No office hours today TA office hours this week as usual Guest lecture Thursday by Suyog Jain Interactive segmentation
[Figure by J. Shi] [http://poseidon.csd.auth.gr/LAB_RESEARCH/Latest/imgs/S peakDepVidIndex_img2.jpg]
Determine image regions Group video frames into shots Fg / Bg
[Figure by Wang & Suter]
Object-level grouping Figure-ground
[Figure by Grauman & Darrell]
Slide credit: Kristen Grauman
Slide credit: Kristen Grauman
http://chicagoist.com/attachments/chicagoist_alicia/GEESE.jpg, http://wwwdelivery.superstock.com/WI/223/1532/PreviewComp/SuperStock_1532R-0831.jpg
Slide credit: Kristen Grauman
http://seedmagazine.com/news/2006/10/beauty_is_in_the_processingtim.php
Slide credit: Kristen Grauman
Image credit: Arthus-Bertrand (via F. Durand)
Slide credit: Kristen Grauman
http://www.capital.edu/Resources/Images/outside6_035.jpg
Slide credit: Kristen Grauman
In Vision, D. Marr, 1982
Slide credit: Kristen Grauman
http://entertainthis.usatoday.com/2015/09/09/how-tom-hardys-legend- poster-hid-this-hilariously-bad-review/
Slide credit: Kristen Grauman
In Vision, D. Marr, 1982; from J. L. Marroquin, “Human visual perception of structure”, 1976.
image human segmentation
Source: Lana Lazebnik
Source: Lana Lazebnik
black pixels gray pixels white pixels
Slide credit: Kristen Grauman
Slide credit: Kristen Grauman
Slide credit: Kristen Grauman
Slide credit: Dan Klein
190 255
intensity
Slide credit: Kristen Grauman
Slide credit: Kristen Grauman
to closest mean
the average of its assigned points
Slide credit: Andrew Moore K-means slides by Andrew Moore Slide credit: Andrew Moore
Slide credit: Andrew Moore Slide credit: Andrew Moore
Slide credit: Andrew Moore Slide credit: Andrew Moore
Source: Steve Seitz
Slide credit: Dan Klein
Slide credit: Kristen Grauman
Slide credit: Steve Seitz
– estimate probability that a point is in a cluster
– mean and variance info (covariance matrix)
Slide credit: Steve Seitz
Slide credit: Kristen Grauman
K=2 K=3
img_as_col = double(im(:)); cluster_membs = kmeans(img_as_col, K); labelim = zeros(size(im)); for i=1:k inds = find(cluster_membs==i); meanval = mean(img_as_column(inds)); labelim(inds) = meanval; end
quantization of the feature space; segmentation label map
Slide credit: Kristen Grauman
R=255 G=200 B=250 R=245 G=220 B=248 R=15 G=189 B=2 R=3 G=12 B=2 R G B
Slide credit: Kristen Grauman
Slide credit: Kristen Grauman
X
Y Intensity Both regions are black, but if we also include position (x,y), then we could group the two into distinct segments; way to encode both similarity & proximity.
Slide credit: Kristen Grauman
F24
F2
F1
Filter bank
statistics to summarize patterns in small windows mean d/dx value mean d/dy value
4 10 Win.#2 18 7 Win.#9 20 20
…
Dimension 1 (mean d/dx value) Dimension 2 (mean d/dy value) Windows with small gradient in both directions Windows with primarily vertical edges Windows with primarily horizontal edges Both
Slide credit: Kristen Grauman
Malik, Belongie, Leung and Shi. IJCV 2001.
Texton map Image
Adapted from Lana Lazebnik
Texton index Texton index Count Count Count Texton index
Slide credit: Kristen Grauman
Slide credit: Kristen Grauman
http://www.caip.rutgers.edu/~comanici/MSPAMI/msPamiResults.html
Space Analysis, PAMI 2002.
Slide credit: Lana Lazebnik
Search window Center of mass Mean Shift vector
Slide by Y . Ukrainitz & B. Sarel
Search window Center of mass Mean Shift vector
Slide by Y . Ukrainitz & B. Sarel
Search window Center of mass Mean Shift vector
Slide by Y . Ukrainitz & B. Sarel
Search window Center of mass Mean Shift vector
Slide by Y . Ukrainitz & B. Sarel
Search window Center of mass Mean Shift vector
Slide by Y . Ukrainitz & B. Sarel
Search window Center of mass Mean Shift vector
Slide by Y . Ukrainitz & B. Sarel
Search window Center of mass
Slide by Y . Ukrainitz & B. Sarel
Slide by Y . Ukrainitz & B. Sarel
http://www.caip.rutgers.edu/~comanici/MSPAMI/msPamiResults.html
q
– wpq measures similarity
» similarity is inversely proportional to difference (in color and position…)
p wpq
Source: Steve Seitz
– similar pixels should be in the same segments – dissimilar pixels should be in different segments
Source: Steve Seitz
q p wpq
Slide credit: Kristen Grauman
σ=.1 σ=.2 σ=1 σ=.2
Source: Steve Seitz
B q A p q p
, ,
[Shi & Malik, 2000 PAMI]
assoc(A,V) = sum of weights of all edges that touch A
with high weights, and few edges of low weight between them
generalized eigenvalue problem.
Source: Steve Seitz
– Dense, highly connected graphs many affinity computations – Solving eigenvalue problem
Slide credit: Kristen Grauman
[Jain & Grauman, HCOMP 2016]