artistic stylization of face photos
play

Artistic Stylization of Face Photos Based on a Single Exemplar - PowerPoint PPT Presentation

Zili Yi Supervisor: Minglun Gong 2015-11-01 Artistic Stylization of Face Photos Based on a Single Exemplar Outline Introduction Background Method Results Conclusion Reference input Introduction example Style Transfer


  1. Zili Yi Supervisor: Minglun Gong 2015-11-01 Artistic Stylization of Face Photos Based on a Single Exemplar

  2. Outline  Introduction  Background  Method  Results  Conclusion  Reference

  3. input Introduction example Style Transfer Stylization of Faces Non-photorealistic Rendering Challenge Region-specific output Viewers’ Sensitivity to face Variation of faces

  4. Style-specific methods Halftoning, Li H. et. al. Pencil drawing, Lu c. et. al. Tessellation, Li h. et. Al. Screening, Li H. et. Al.

  5. Data driven methods Style transfer, Shih Y. et. al. Guided Transfer, Li H. Data Driven Cartoonizing, Zhang Y. et. al.

  6. Our Task Data Driven Style-Specific General stylization with a single Example

  7. Background  Face Detection & Alignment  Non-Photo Realistic Rendering  Color Style Transfer  Optimized Texture Synthesis

  8. Framework of Our Method Color Texture Transfer Transfer Example

  9. Framework of Our Method Color Texture Transfer Transfer Example

  10. Framework of Our Method Color Texture Transfer Transfer Example

  11. Color Transfer Color Texture Transfer Transfer Example

  12. Color Transfer 320*448 image Face face detection& landmark normalization label assignment color dense assignment correspondence

  13. Texture Transfer Color Texture Transfer Transfer Example

  14. Texture Transfer  We view the problem of Texture Transfer as Edge- Preserved Texture synthesis problem  Optimized Texture synthesis  Iteratively optimizing and voting

  15. Optimizing  Finding the Nearest Neighbor Field  Generalized PatchMatch S ‘ measures the L 2 distance Between patch and E

  16. Voting  Simply averaging the Nearest Neighbors

  17. Pyramid  The texture is synthesized in a pyramid manner  Coarse to fine  10 scales (ratio 1.5)

  18. Preserve the edges  Pyramid Edge Map  Using edge maps as the preserving mask

  19. The final synthesized image

  20. Experiment Result

  21. Results

  22. Comparison

  23. Comparison

  24. Conclusion  Strength  Robust to gender, skin color, hair style, face accessories, beards, glasses, slight variation in poses, lightening condition  Able to transfer wide varieties of styles (pencil drawing, sand drawing, oil painting, mosaic, screening, water color painting, Chinese painting and Pyrography)  Limitation  Time consumption (over 40 mins to process one image)  Unable to handle stylization which involves geometry exaggeration  T end to blur textures with long curve patterns.

  25. Reference

  26. Reference

Recommend


More recommend