1 / 46

Image Processing on Line A new way to publish ?

Image Processing on Line A new way to publish ? A new way to organize research in a lab ? Jérôme Darbon (UCLA and ENS Cachan), Nicolas Limare (ENS Cachan), Jean-Michel Morel (ENS Cachan). Research partially funded by ONR Grant N00014-97-1-0839.

amish
Download Presentation

Image Processing on Line A new way to publish ?

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. Image Processing on Line A new way to publish? A new way to organizeresearch in a lab? Jérôme Darbon (UCLA and ENS Cachan), Nicolas Limare (ENS Cachan), Jean-Michel Morel (ENS Cachan) Researchpartiallyfunded by ONR Grant N00014-97-1-0839 Image Processing on Line, Minneapolis, ONR meeting 2009

  2. Image Processing on Line: main goal Achieve « reproducibleresearch » and therefore: Allow EVERYBODY to try the algorithms on theirown images (includingdeciders who DO NOT program) Make all results of trials accessible in a Web archive (experiment sharing) Downloadable code Give a list of examples AND counterexamples, and explainthem Describe the algorithmscarefully (pseudocode+comments) Testingindependent of anyplatform (no code download, no system requirement…) This isparticularlyadapted to image processing! Image Processing on Line, Minneapolis, ONR meeting 2009

  3. Means of the project Online servers, multicore Execution in real/interactive time (lessthan 20 seconds) Research team on online web tools Some (easy) parallelcomputing Reorganization of the work flow in a team Image Processing on Line, Minneapolis, ONR meeting 2009

  4. Reorganization of the work flow in a research team Image Processing on Line, Minneapolis, ONR meeting 2009

  5. Four functions for a Web server Image Processing on Line, Minneapolis, ONR meeting 2009

  6. For eachalgorithm: a fourfold publication Image Processing on Line, Minneapolis, ONR meeting 2009

  7. Current state of the site PUBLIC ONLINE (Website and demos): http://mw.cmla.ens-cachan.fr/megawave/algo/: ASIFT : Affine Invariant Image Matching LSD : Line Segment Detection Random Phase Noise : MicrotextureSynthesis Cartoon+Texture Image Decomposition Retinex PDE Color Balance INTERNAL WORKSHOPS: Colorcontrastenhancement by PDE (threealgorithmscompared) http://mw.cmla.ens-cachan.fr/megawave/demo/retinex_workshop/ NL-means (on test) http://mw.cmla.ens-cachan.fr/megawave/demo/nlmeans/ Image Processing on Line, Minneapolis, ONR meeting 2009

  8. ASIFT: affine invariant image comparison Image Processing on Line, Minneapolis, ONR meeting 2009

  9. Result of ASIFT: Comparedwith SIFT: Image Processing on Line, Minneapolis, ONR meeting 2009

  10. AlgorithmWebsite Image Processing on Line, Minneapolis, ONR meeting 2009

  11. The website of eachalgorithms shows and explains the failure cases. For instance for SIFT and ASIFT, failurecomparingobjectswith night and day illumination Image Processing on Line, Minneapolis, ONR meeting 2009

  12. The online demogivesalsoaccess to the online archive. More than 1550 different images have been so far tried by on line users. They are grouped in pages of 50. Here are three examplestried by users, on a simple box, a building and a landscape. Image Processing on Line, Minneapolis, ONR meeting 2009

  13. By a simple click a closerview of eachexperimentisavailable: ASIFT result : SIFT result : Image Processing on Line, Minneapolis, ONR meeting 2009

  14. Image Processing on Line, Minneapolis, ONR meeting 2009

  15. Line segment detector (LSD), no parameter. More than 850 images in the archive Image Processing on Line, Minneapolis, ONR meeting 2009

  16. Someresults Image Processing on Line, Minneapolis, ONR meeting 2009

  17. LSD: Someresults in the online archive Image Processing on Line, Minneapolis, ONR meeting 2009

  18. Image Processing on Line, Minneapolis, ONR meeting 2009

  19. Image Processing on Line, Minneapolis, ONR meeting 2009

  20. Image Processing on Line, Minneapolis, ONR meeting 2009

  21. Image Processing on Line, Minneapolis, ONR meeting 2009

  22. Image Processing on Line, Minneapolis, ONR meeting 2009

  23. Image Processing on Line, Minneapolis, ONR meeting 2009

  24. Image Processing on Line, Minneapolis, ONR meeting 2009

  25. Image Processing on Line, Minneapolis, ONR meeting 2009

  26. Image Processing on Line, Minneapolis, ONR meeting 2009

  27. Plans • Make a substantialdatabase of algorithms • . New ones • . But manyclassicones • Passfromatoms to molecules : • . Stereo (calibration, stereorectification, matching, reconstruction • . Image processing (noise, blur, color, contrast: requiresreliableestimates) • Official publication? • . Needs a scientific AND technicalcommittee • . Needs a union of labs to start (right nowthreelabsvolunteer) • . Needs to fixverycarefully the rules for code description, examples, etc. • . Needs €€€€€ + $$$$$ • VISIT US: http://mw.cmla.ens-cachan.fr/megawave/algo/: Image Processing on Line, Minneapolis, ONR meeting 2009

  28. Image Processing on Line, Minneapolis, ONR meeting 2009

  29. Image Processing on Line, Minneapolis, ONR meeting 2009

  30. Image Processing on Line, Minneapolis, ONR meeting 2009

  31. Image Processing on Line, Minneapolis, ONR meeting 2009

  32. Image Processing on Line, Minneapolis, ONR meeting 2009

  33. Image Processing on Line, Minneapolis, ONR meeting 2009

  34. Image Processing on Line, Minneapolis, ONR meeting 2009

  35. Image Processing on Line, Minneapolis, ONR meeting 2009

  36. Image Processing on Line, Minneapolis, ONR meeting 2009

  37. Image Processing on Line, Minneapolis, ONR meeting 2009

  38. Image Processing on Line, Minneapolis, ONR meeting 2009

  39. Image Processing on Line, Minneapolis, ONR meeting 2009

  40. Image Processing on Line, Minneapolis, ONR meeting 2009

  41. Image Processing on Line, Minneapolis, ONR meeting 2009

  42. Image Processing on Line, Minneapolis, ONR meeting 2009

  43. Image Processing on Line, Minneapolis, ONR meeting 2009

  44. Image Processing on Line, Minneapolis, ONR meeting 2009

  45. Image Processing on Line, Minneapolis, ONR meeting 2009

  46. Image Processing on Line, Minneapolis, ONR meeting 2009

More Related