1 / 24

Inpainting Algorithms for Image Coding and Scaling

Inpainting Algorithms for Image Coding and Scaling. Patrick Ndjiki-Nya. Outline. Motivation Image Inpainting and Coding Structure Interpolation Post-Processing Methods Results Image Inpainting and Spatial Scaling Proposed Method Results Conclusions and Future Work. Motivation.

akira
Download Presentation

Inpainting Algorithms for Image Coding and Scaling

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. Inpainting Algorithms for Image Coding and Scaling Patrick Ndjiki-Nya

  2. Outline • Motivation • Image Inpainting and Coding • Structure Interpolation • Post-Processing Methods • Results • Image Inpainting and Spatial Scaling • Proposed Method • Results • Conclusions and Future Work

  3. Motivation High-frequency textures are difficult to code, e.g. flower bed, crushed stones, water etc. Where is the original texture ?

  4. Coding: Block Diagram MSE Data Encoder Decoder Video In Video Out Synth. Metadata TS TAS Texture Analysis-Synthesis (TAS) Video Coding

  5. Coding: Synthesizer Input Current frame (t0)

  6. Right ref. frame (t0+x) Left ref. frame (t0-x) Current frame (t0) Coding: Rigid Texture Synthesizer (TSI) • Texture synthesizer for rigid (flower bed, sand, …) textures • Parametric approach • Features some similarity with global motion compensation • Typical drawback: Limited by assumptions of motion model

  7. Coding: Non-Rigid Texture Synth. (TSII, 2D) Stuffing procedure • Generic approach: Good for rigid (flower bed, sand, …) as well as non-rigid textures (water, smoke,…) • Non-parametric approach • Typical drawback: Complexity of texture representation

  8. Coding: Non-Rigid Texture Synth. (TSII, 2D+t)

  9. Coding: Intermediate Results

  10. Coding: Intermediate Results (cont.)

  11. Coding: Challenges

  12. Coding: Structure Interpolation

  13. Coding: Structure Interpolation

  14. Coding: Structure Interpolation Results

  15. Coding: Structure Interpolation Results

  16. Coding: Structure Interpolation Results

  17. Simple Cloning Poisson Seamless Cloning Post-Processing Methods: Poisson Cloning

  18. v g Post-Processing Methods: Formalization • S: a closed subset of R2 • Ω: a closed subset of S, with boundary ∂Ω • f*: known scalar function over S\Ω • f: unknown scalar function over Ω • ∂Ω: boundary pixels • g: source function • v: vector field of g

  19. with over Ω with Post-Processing Methods: Formalization Poisson equation with Dirichlet boundary condition

  20. Coding: Structure Interpolation Results

  21. Image Inpainting and Small Holes

  22. Structure-aware Interpolation Bicubic Interpolation Scaling Results

  23. Conclusions and Future Work • Efficient tools for image inpainting presented • Performance optimization of cloning in inpainting framework • Extension of structure-aware methods to video (rigid and non-rigid textures)

  24. Thank You

More Related