Presentation is loading. Please wait.

Presentation is loading. Please wait.

Computer Vision Fitting Marc Pollefeys COMP 256 Some slides and illustrations from D. Forsyth, T. Darrel, A. Zisserman,...

Similar presentations


Presentation on theme: "Computer Vision Fitting Marc Pollefeys COMP 256 Some slides and illustrations from D. Forsyth, T. Darrel, A. Zisserman,..."— Presentation transcript:

1 Computer Vision Fitting Marc Pollefeys COMP 256 Some slides and illustrations from D. Forsyth, T. Darrel, A. Zisserman,...

2 Computer Vision Aug 26/28-Introduction Sep 2/4CamerasRadiometry Sep 9/11Sources & ShadowsColor Sep 16/18Linear filters & edges(hurricane Isabel) Sep 23/25Pyramids & TextureMulti-View Geometry Sep30/Oct2StereoProject proposals Oct 7/9Tracking (Welch)Optical flow Oct 14/16-- Oct 21/23Silhouettes/carving(Fall break) Oct 28/30-Structure from motion Nov 4/6Project updateProj. SfM Nov 11/13Camera calibrationSegmentation Nov 18/20FittingProb. segm.&fit. Nov 25/27Matching templates(Thanksgiving) Dec 2/4Matching relationsRange data Dec 8 or 9?Final project Tentative class schedule

3 Computer Vision Final project presentation Presentation and/or Demo (your choice, but let me know) Short paper (Due Dec.5) Final presentation/demo Monday 8, 2-5pm?

4 Computer Vision Last week: Segmentation Group tokens into clusters that fit together –foreground-background –cluster on intensity, color, texture, location, … K-means graph-based

5 Computer Vision Fitting Choose a parametric object/some objects to represent a set of tokens Most interesting case is when criterion is not local –can’t tell whether a set of points lies on a line by looking only at each point and the next. Three main questions: –what object represents this set of tokens best? –which of several objects gets which token? –how many objects are there? (you could read line for object here, or circle, or ellipse or...)

6 Computer Vision Fitting and the Hough Transform Purports to answer all three questions –in practice, answer isn’t usually all that much help We do for lines only A line is the set of points (x, y) such that Different choices of , d>0 give different lines For any (x, y) there is a one parameter family of lines through this point, given by Each point gets to vote for each line in the family; if there is a line that has lots of votes, that should be the line passing through the points

7 Computer Vision tokens votes

8 Computer Vision Mechanics of the Hough transform Construct an array representing , d For each point, render the curve (, d) into this array, adding one at each cell Difficulties –how big should the cells be? (too big, and we cannot distinguish between quite different lines; too small, and noise causes lines to be missed) How many lines? –count the peaks in the Hough array Who belongs to which line? –tag the votes Hardly ever satisfactory in practice, because problems with noise and cell size defeat it

9 Computer Vision tokens votes

10 Computer Vision

11 Computer Vision

12 Computer Vision

13 Computer Vision Cascaded hough transform Tuytelaars and Van Gool ICCV‘98

14 Computer Vision Line fitting can be max. likelihood - but choice of model is important

15 Computer Vision Who came from which line? Assume we know how many lines there are - but which lines are they? –easy, if we know who came from which line Three strategies –Incremental line fitting –K-means –Probabilistic (later!)

16 Computer Vision

17 Computer Vision

18 Computer Vision

19 Computer Vision

20 Computer Vision

21 Computer Vision

22 Computer Vision

23 Computer Vision

24 Computer Vision

25 Computer Vision

26 Computer Vision

27 Computer Vision

28 Computer Vision

29 Computer Vision

30 Computer Vision Robustness As we have seen, squared error can be a source of bias in the presence of noise points –One fix is EM - we’ll do this shortly –Another is an M-estimator Square nearby, threshold far away –A third is RANSAC Search for good points

31 Computer Vision

32 Computer Vision

33 Computer Vision

34 Computer Vision

35 Computer Vision M-estimators Generally, minimize whereis the residual

36 Computer Vision

37 Computer Vision

38 Computer Vision

39 Computer Vision Too small

40 Computer Vision Too large

41 Computer Vision

42 Computer Vision RANSAC Choose a small subset uniformly at random Fit to that Anything that is close to result is signal; all others are noise Refit Do this many times and choose the best Issues –How many times? Often enough that we are likely to have a good line –How big a subset? Smallest possible –What does close mean? Depends on the problem –What is a good line? One where the number of nearby points is so big it is unlikely to be all outliers

43 Computer Vision

44 Computer Vision Distance threshold Choose t so probability for inlier is α (e.g. 0.95) Often empirically Zero-mean Gaussian noise σ then follows distribution with m =codimension of model (dimension+codimension=dimension space) CodimensionModelt 2 1line,F3.84σ 2 2H,P5.99σ 2 3T7.81σ 2

45 Computer Vision How many samples? Choose N so that, with probability p, at least one random sample is free from outliers. e.g. p=0.99 proportion of outliers e s5%10%20%25%30%40%50% 2235671117 33479111935 435913173472 54612172657146 64716243797293 748203354163588 8592644782721177

46 Computer Vision Acceptable consensus set? Typically, terminate when inlier ratio reaches expected ratio of inliers

47 Computer Vision Adaptively determining the number of samples e is often unknown a priori, so pick worst case, e.g. 50%, and adapt if more inliers are found, e.g. 80% would yield e =0.2 –N=∞, sample_count =0 –While N >sample_count repeat Choose a sample and count the number of inliers Set e=1-(number of inliers)/(total number of points) Recompute N from e Increment the sample_count by 1 –Terminate

48 Computer Vision Step 1. Extract features Step 2. Compute a set of potential matches Step 3. do Step 3.1 select minimal sample (i.e. 7 matches) Step 3.2 compute solution(s) for F Step 3.3 determine inliers until  (#inliers,#samples)<95% #inliers90%80%70%60%50% #samples51335106382 Step 4. Compute F based on all inliers Step 5. Look for additional matches Step 6. Refine F based on all correct matches (generate hypothesis) (verify hypothesis) RANSAC for Fundamental Matrix

49 Computer Vision Step 1. Extract features Step 2. Compute a set of potential matches Step 3. do Step 3.1 select minimal sample (i.e. 7 matches) Step 3.2 compute solution(s) for F Step 3.3 Randomize verification 3.3.1 verify if inlier while hypothesis is still promising while  (#inliers,#samples)<95% Step 4. Compute F based on all inliers Step 5. Look for additional matches Step 6. Refine F based on all correct matches (generate hypothesis) (verify hypothesis) Randomized RANSAC for Fundamental Matrix

50 Computer Vision Example: robust computation Interest points (500/image) (640x480) Putative correspondences (268) (Best match,SSD<20,±320) Outliers (117) (t=1.25 pixel; 43 iterations) Inliers (151) Final inliers (262) (2 MLE-inlier cycles; d  =0.23→d  =0.19; Iter Lev-Mar =10) #in 1-e adapt. N 62%20M 103%2.5M 4416%6,922 5821%2,291 7326%911 15156%43 from H&Z

51 Computer Vision More on robust estimation LMedS, an alternative to RANSAC (minimize Median residual in stead of maximizing inlier count) Enhancements to RANSAC –Randomized RANSAC –Sample ‘good’ matches more frequently –… RANSAC is also somewhat robust to bugs, sometimes it just takes a bit longer…

52 Computer Vision Epipolar geometry from silhouettes RANSAC is used to combine exploration of random epipole locations and robustness to outliers (Sinha et al. CVPR 2004? paper due tomorrow…) more on Dec. 8 …

53 Computer Vision Fitting curves other than lines In principle, an easy generalisation –The probability of obtaining a point, given a curve, is given by a negative exponential of distance squared In practice, rather hard –It is generally difficult to compute the distance between a point and a curve

54 Computer Vision Next class: Segmentation and Fitting using Probabilistic Methods Reading: Chapter 16 Missing data: EM algorithm Model selection


Download ppt "Computer Vision Fitting Marc Pollefeys COMP 256 Some slides and illustrations from D. Forsyth, T. Darrel, A. Zisserman,..."

Similar presentations


Ads by Google