1 / 28

Andrei Sharf Dan A. Alcantara Thomas Lewiner Chen Greif

Space-time Surface Reconstruction using Incompressible Flow. Andrei Sharf Dan A. Alcantara Thomas Lewiner Chen Greif Alla Sheffer Nina Amenta Daniel Cohen-Or. 4D Data acquisition - setting. Synchronized static cameras 2-16 views Capture rates 10-30 fps.

morna
Download Presentation

Andrei Sharf Dan A. Alcantara Thomas Lewiner Chen Greif

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Space-time Surface Reconstruction using Incompressible Flow Andrei Sharf Dan A. Alcantara Thomas Lewiner Chen Greif Alla Sheffer Nina Amenta Daniel Cohen-Or

  2. 4D Data acquisition - setting • Synchronized static cameras • 2-16 views • Capture rates 10-30 fps

  3. 4D Data acquisition - limitations • Persistent self occlusions • Low frame rate and resolution • Noise

  4. time mass field 2D Motivation • Volume is incompressible across time • Explicitly modeling of mass field: • Compute inside/outside volume • Include physical assumptions • Object is watertight manifold

  5. Contribution • Incompressible mass flow 4D reconstruction • Global system considers all frames simultaneously • Simple formulation on a grid • General un-constrained deformations time

  6. Related work Guskov et al. 2003 • Marker based[Marschner et al. 2000; Guskov et al. 2003; White et al. 2007] • Template based[Allen et al. 2002; Anguelov et al. 2004; Zhang et al. 2004; Anguelov et al. 2005, Aguiar et al. 2008; Bradley et al. 2008] • Point correspondence and registration[Shinya 2004; Wang et al. 2005, Anuar and Guskov 2004, Pekelny and Gotsman 2008, Chang and Zwicker 2008; Li et al. 2008] • Surface based space-time[Wand et al. 2007; Mitra et al. 2007; Suessmuth et al. 2008] Zhang et al. 2004 Anuar et al. 2004 Li et al. 2008 Mitra et al. 2007

  7. FLOW reconstruction • In: 3D point cloud frames • Out: watertight surface • Explicit volume modeling • 4D solid on a grid • Characteristic function

  8. 3D reconstruction techniques fail • Surface reconstruction of individual frames fails

  9. known outside 0 known inside 1 unknown [0-1] 2D example

  10. 1D representation • Domain: space time grid • Material: characteristic function xti values mass amount at each cell • Flow: amount of material vti,j moving from cell xti to cell xt+1j xt+1i-1 xt+1i xt+1i+1 t+1 vti,i vti,i-1 vti,i+1 time xti t

  11. Higher dimensions generalization • Regular 4D grid on top of 3D scan frames • Space-time adjacency relationships: 1D 2D 3D

  12. FLOW physical constraints • Mass preservation: material in cell equals to material flowing into and out of the cell time

  13. FLOW physical constraints • Spatial continuity: values spatially adjacent to be identical everywhere, except across boundaries space

  14. FLOW Physical Constraints • Flow momentum: flow direction should be smooth across time

  15. Constrained Minimization Problem Optimization: Constraints: Incompressibility constraints Boundary values

  16. Challenges Sublinear exponent iterative reweighted least squares Huge matrices fine-tuned iterative solver Mass stability boundary constraints, clamping

  17. Sublinear exponent Iteratively Reweighted Least Squares: from previous iteration small close to discontinuities converges with good init and few outliers time iteration

  18. Huge data problem Problem size 20-200 frames x (28)3 grid resolution x 8 variables per cell in time Reduce initial number of unknowns Pre-assignment from visibility hull : inside/outside labels High resolution per-frame surface reconstruction Sharf et al. 07

  19. Lagrange multipliers: Minimization problem

  20. Matrix engineering Iterative, preconditioned with many eigenvalues 1fast convergence Augmenting approach:solve with CGMINRES solver with decreasing tolerance

  21. Mass stability: clamping/back substitution : amount of mass at cell (i,t) : inside, outside clamp if then adjacent back-substitution reduces the system size time iteration

  22. Results (2D) – empty frames completion

  23. Results – hand puppet from 2 views

  24. Results - garments

  25. Results – large deformation

  26. Results 3D+time • 20 frames at constant resolution • Solver converges in 100 iterations. • Time: 1 minute per-frame • 3.73 GHz CPU, memory requirements up to 4.5GB

  27. Thank you!

More Related