Exampled-based Super resolution - PowerPoint PPT Presentation

exampled based super resolution n.
Download
Skip this Video
Loading SlideShow in 5 Seconds..
Exampled-based Super resolution PowerPoint Presentation
Download Presentation
Exampled-based Super resolution

play fullscreen
1 / 22
Exampled-based Super resolution
242 Views
Download Presentation
ilar
Download Presentation

Exampled-based Super resolution

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. Exampled-based Super resolution Presenter: Yu-Wei Fan

  2. Outline • Introduction • Training set generation • Super-resolution algorithms • Idea • Markov Network • One-pass algorithm • Results

  3. Outline • Introduction • Training set generation • Super-resolution algorithms • Idea • Markov Network • One-pass algorithm • Results

  4. Introduction • Why do we need high resolution image? • Usually , we cannot get high resolution image easy.

  5. Introduction • Aim: High Resolution Image • 1.Reduce the pixel size • the amount of light available also decrease • generates shot noise • 2.Increase the chip size • increase capacitance • difficult to speed up a charge transfer rate • 3.Signal processing techniques • Low cost

  6. Introduction • General Super Resolution • Need multi frames information • Exampled-based Super resolution • Need only one frame

  7. Outline • Introduction • Training set generation • Super-resolution algorithms • Idea • Markov Network • One-pass algorithm • Results

  8. Training set generation • Store the high-resolution patch corresponding to every possible • low-resolution image patch. • Typically, these patches are 5 × 5 or 7 × 7 pixels.

  9. Outline • Introduction • Training set generation • Super-resolution algorithms • Idea • Markov Network • One-pass algorithm • Results

  10. Idea Unfortunately, that approach doesn’t work!

  11. Markov Network

  12. Markov Network • MAP Estimator:

  13. Markov Network • Example:

  14. Markov Network • Belief Propagation Where is from the previous iteration. The initial are 1. Typically, three or four iterations of the algorithm are sufficient.

  15. One-pass algorithm • How do we select a good patch pair? • Two constraint: • frequency constraint • spatial constraint

  16. One-pass algorithm

  17. Outline • Introduction • Training set generation • Super-resolution algorithms • Idea • Markov Network • One-pass algorithm • Results

  18. Results

  19. Results

  20. Results • α=0

  21. Results • α=0.5

  22. Results • α=5