a gentle introduction a gentle introduction to bilateral
play

A Gentle Introduction A Gentle Introduction to Bilateral Filtering - PowerPoint PPT Presentation

A Gentle Introduction A Gentle Introduction to Bilateral Filtering to Bilateral Filtering and its Applications and its Applications Sylvain Paris MIT CSAIL Pierre Kornprobst INRIA Odysse Jack Tumblin Northwestern University


  1. A Gentle Introduction A Gentle Introduction to Bilateral Filtering to Bilateral Filtering and its Applications and its Applications Sylvain Paris – MIT CSAIL Pierre Kornprobst – INRIA Odyssée Jack Tumblin – Northwestern University Frédo Durand – MIT CSAIL

  2. • The bilateral filter is becoming in computational photography. • Many applications with high quality results.

  3. Photographic Style Transfer Photographic Style Transfer [Bae 06] [Bae 06] input

  4. Photographic Style Transfer Photographic Style Transfer [Bae 06] [Bae 06] output

  5. Tone Mapping Tone Mapping [Durand 02] [Durand 02] HDR input

  6. Tone Mapping Tone Mapping [Durand 02] [Durand 02] output

  7. input Cartoon Rendition Cartoon Rendition [Winnemöller 06] [Winnemöller 06]

  8. Cartoon Rendition Cartoon Rendition [Winnemöller 06] [Winnemöller 06] 6 papers at 6 papers at SIGGRAPH’07 SIGGRAPH’07 output

  9. Goal: Image Smoothing Goal: Image Smoothing Split an image into: • large-scale features, structure • small-scale features, texture

  10. Naïve Approach: Gaussian Blur Naïve Approach: Gaussian Blur BLUR HALOS smoothed residual input ( structure, large scale ) ( texture, small scale ) Gaussian Convolution

  11. Impact of Blur and Halos Impact of Blur and Halos • If the decomposition introduces blur and halos, the final result is corrupted. Sample manipulation: increasing texture (residual × 3)

  12. Bilateral Filter: no Blur, no Halos Bilateral Filter: no Blur, no Halos smoothed residual input ( structure, large scale ) ( texture, small scale ) edge-preserving: Bilateral Filter

  13. input

  14. increasing texture with Gaussian convolution H A L O S

  15. increasing texture with bilateral filter N O H A L O S

  16. Many Other Options Many Other Options • Bilateral filtering is not the only image smoothing filter – Diffusion, wavelets, Bayesian… • We focus on bilateral filtering – Suitable for strong smoothing used in computational photography – Conceptually simple

  17. Content of the Course Content of the Course All you need to know about bilateral filtering: – Definition of the bilateral filter – Parameter influence and settings – Applications – Relationship to other filters – Theoretical properties – Efficient implementation

  18. Course Material Course Material • Course webpage (google “bilateral filter course”): http://people.csail.mit.edu/sparis/siggraph07_course/ – Detailed course notes – Slides (soon) – C++ and Matlab code – Links

  19. A Gentle Introduction A Gentle Introduction to Bilateral Filtering and its Applications to Bilateral Filtering and its Applications • From Gaussian blur to bilateral filter – S. Paris • Applications – F. Durand • Link with other filtering techniques – P. Kornprobst BREAK • Implementation – S. Paris • Variants – J. Tumblin • Advanced applications – J. Tumblin • Limitations and solutions – P. Kornprobst

Recommend


More recommend