1. Unwrap Mosaics: A new representation for video editing Alex Rav-Acha et al. In SIGGRAPH 2008 발표 이성호 2009 년 1 월 22 일

2. Abstract A new representation for video Modeling the image-formation process –From an object’s texture map to the image Modulated by an object-space occlusion mask –Recover “unwrap mosaic” Editing Re-composition –Into the original space –Resizing objects –Repainting textures –Copying/cutting/pasting objects –Attaching effects layers to deforming objects 2

3. Introduction Unwrap mosaics –A wide range of Editing –Deforming surface –Self-occlusion –Provides references for feature-point tracking 3

4. Reconstructiong 3D model Not easy Software packages –[2d3 Ltd. 2008; Thorm¨ahlen and Broszio 2008] Extensions to nonrigid scenes –[Bregler et al. 2000; Brand 2001; Torresani et al. 2008] Less reliable under occlusion 4

5. Dense reconstruction from video Restricted to rigid scenes –[Seitz et al. 2006] Using interactive tools –[Debevec et al. 1996; van den Hengel et al. 2007] –Also for rigid scenes Triangulation of the sparse points from nonrigid structure –Surprisingly troublesome 5

6. Unwrap mosaics Ro recover the object’s texture map –Rather than its 3D shape –Directly from video recovered texture map will be a –2D-to-2D mapping –Sequence of binary masks modeling occlusion –Unwrap mosaic 6

7. A video will typically be represented –by an assembly of several unwrap mosaics one per object, and one for the background. Edits –performed on mosaic itself –Without converting to 3D Main contribution –Recover the unwrap mosaic from images Energy minimization procedure 7

8. Figure 2: Reconstruction overview. Steps in constructing an unwrap mosaic representation of a video sequence. Steps 1 to 3a form an initial estimate of the model parameters, and step 3 is an energy minimization procedure which refines the model to subpixel accuracy. 8

9. Segmentation Segment the sequence into –independently moving objects “video cut and paste” [Li et al. 2005] Allow for user interaction 9

10. Tracking Recover the texturemap of –a deforming 3D object –from a sequence of 2D images texture map may be assumed to be constant –although the model is changing its shape Interestpoint detection and tracking –[Sand and Teller 2006, for example] 10

11. Embedding view the sparse point tracks –as a high dimensional projection of the 2D surface parameters 11

12. Mosaic stitching a map from the tracked points in each image –to the (u; v) parameter space. A variation of [Agarwala et al. 2004] –emerges naturally from the energy formulation. 12

13. Track refinement the mosaic is good enough –to create a reference template to match against the original frames reduces any drift that may have been present after the original tracking phase 13

14. Using the model for video editing Edit the texture map –For example by drawing on it –Warp it via the recovered mapping –Combine with the other layers of the original sequence Re-rendered mosaic will not exactly match the original sequence –warped by the 2D–2D mapping –masked by the occlusion masks –alpha-blended with the original image Remove layers –The removed area is filled in because the mapping is defined even in occluded areas. 14

15. Limitations Textured surfaces are required for point tracking –Low texture –One dimensional textures –motion blur The assumption of a smoothly varying smooth 3D surface –objects with significant protrusions the dinosaur in figure 11 The assumption of smoothly varying lighting –strong shadows will disrupt tracking limited to disc-topology objects –rotating cylinder will be reconstructed as a long tapestry see figure 11 15

16. The unwrap mosaic model Image generation model –how an image sequence is constructed from a collection of unwrap mosaics –a fitting problem Energy minimization –via nonlinear optimization Initial estimate for the –2D-2D mapping –Texture map –Can be obtained from sparse 2D tracking data. 16

17. 17

18. 18

19. Point-spread functions 19

20. 20

21. 21

22. 22

23. 23

24. 24

25. Discrete energy formulation 25

26. 26

27. Data cost 27

28. 28

29. Constraints 29

30. Mapping smoothness 30

31. Visibility smoothness 31

32. Minimizing the energy 32

33. Minimizing over C: stitching 33

34. 34

35. 35

36. Reparametrization and embedding 36

37. 37

38. 38

39. Minimizing over w: dense mapping Simply use MATLAB’s griddata to interpolate –No guarantee of minimizing the original energy 39

40. Minimizing over w and b: dense mapping with occlusion 40

41. 41

42. Lighting 42

43. User interaction and hinting Segmentatinon –the user selects objects in a small number of frames, –and the segmentation is propagated using optical flow or sparse tracks. mosaic coverage 43

44. 44

45. Tuning parameters The robust kernel width τ –is set to match an estimate of image noise. –Set to 5/255 gray-levels Scale parameter τ3 –40 pixels Except the face, which had many outlier tracks spatial smoothness λwl –controls the amount of deformation of the mapping –Constant for all but the “boy” sequence 45

46. Results Synthetic sequence 46

47. Synthetic sequence There is no concept of a “ground truth” –visually evaluate the recovered mosaic (figure 7) About 30% of mosaic pixels visible –in any one frame Self occlusion near the nose 47

48. Face sequence Few high-contrast points coupled –with strong lighting 48

49. Giraffe sequence: foreground A logo –is placed on the foreground giraffe’s back and head With optical flow, –the annotation drifts by about 10 pixels in 30 frames, while the Unwrap mosaic –shows no visible drift. 49

50. Boy sequence –both sides of his head and torso are shown –variable focus –motion blur –considerable foreground occlusion Ear is doubled –could be fixed by editing as in section 4 50

51. Dinosaur sequence Not a smooth reparametrization –of the object’s true “texture map” iterative automatic segmentation –Might separate mosaics for each model component: arms, torso, tail Unwrap mosaic technique is –deformation agnostic 51

52. Related work Wang and Adelson’s paper [1994] –about how layers might apply to self-occludng objects Irani et al. [1995] –more elaborate parametric transformations Layered Depth Images [Shade et al. 1998] Still photos to mosaics [Bhat et al. 2007] –for rigid scenes Optic flow –energy-based formulation [Bruhn et al. 2005] 52

53. Related work Point tracking –Sand and Teller [2006] Fitting motion models –corresponding to multiple 3D motions Bhat et al. [2007] Discover texture coordinates –for predefined 3D surface models Zigelman et al. [2002] Use of energy formulations –to regularize models –and to make modelling assumptions coherent and explicit [Fleet et al. 2002, Brox et al. 2004, Bruhn et al. 2005] 53

54. Discussion Without 3D shape recovery –manipulations of the video which respect the three-dimensionaliy of the scene Viewing reconstruction –as an embedding of tracks into 2D Embedding would be unstable –For more constrained models –Large amounts of missing data that self-occlusion causes We use approximate algorithms –No guarantees of globally minimizing the overall energy 54

55. Generalizations Automatically segment the layers –Many existing methods could be used –Sparse point tracks are clustered into independent motions Optimizing each layer simultaneously To allow non-boolean masks b Apply matrix factorization to the input tracks –to recover a deformable 3D shape –can place constraints on the mapping which would allow outlier removal –allow exact visibility to be computed 55