1 / 32

Real time Color- FilteredAperture Depth Estimation with single image

Real time Color- FilteredAperture Depth Estimation with single image . Outline. Color Filtered Aperture Depth Estimation Box Filter Matlab C Code With Intel IPP Library Compute Shader Depth E stimation Result. Color filtered aperture. Color filtered aperture Depth estimation.

percy
Télécharger la présentation

Real time Color- FilteredAperture Depth Estimation with single image

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. Real time Color-FilteredAperture Depth Estimation with single image

  2. Outline • Color Filtered Aperture • Depth Estimation • Box Filter • Matlab • C Code With Intel IPP Library • Compute Shader • Depth EstimationResult

  3. Color filtered aperture • Color filtered aperture • Depth estimation

  4. Color filtered aperture Canon EF 50mm f/1.8 II lens Aperture part of the disassembled lens Theprototype lens with color-filters FujifilterSC-58, BPB-53, and BPB-45

  5. Capture Picture

  6. Red Plane

  7. Green Plane

  8. Blue Plane

  9. The Misalignment of RGB

  10. Explain The Color Misaligned Color filters Background Foreground object Image sensor Lens Color filters Background Image sensor Lens

  11. Depth estimation • Color filtered aperture • Depth estimation

  12. Original Picture (disparity = 0)

  13. Disparity =1

  14. Disparity =2

  15. Disparity =3

  16. Disparity =4

  17. Disparity =5

  18. Disparity =6

  19. Color-Alignment Measure • Color alignment metric • where are the eigenvalues of the covariance matrix of the color distribution. • According to PCA (Principle Component Analysis )

  20. Color-Alignment Measure

  21. Color-Alignment Measure • Complicated computation

  22. Explain the complexity R G Disparity=0 Disparity=1 Disparity=2 Disparity=maxD B

  23. Box Filter • Using box filter method to calculate the average of all blocks Simultaneously. 4 = 9 - 3 - 3 + 1

  24. Box Filter

  25. Depth Estimation Result

  26. Depth Estimation Result Background (far) Foreground object Background (near)

  27. C Code with Intel IPP Library • We also use Intel IPP(Integrated Performance Primitives)library to reduce the running time . • Intel IPP is an extensive library of multicore-ready, highly optimized software functions for digital media and data-processing applications. • It runs faster while using Intel IPP library because the IPP library uses SIMD (Single Instruction Multiple Data) instructions.

  28. Compute Shader • A compute shader provides high-speed general purpose computing and takes advantage of the large numbers of parallel processors on the graphics processing unit (GPU).  • A compute shader is designed and implemented with HLSL.

  29. Software And Hardware Environment

  30. languages

  31. Elapsed time

More Related