1 / 47

ICCV 2005 Beijing, Short Course, Oct 15

ICCV 2005 Beijing, Short Course, Oct 15. Recognizing and Learning Object Categories. Li Fei-Fei, UIUC Rob Fergus, MIT Antonio Torralba, MIT. Agenda. Introduction Bag of words models Part-based models Discriminative methods Segmentation and recognition Conclusions. material thing.

Download Presentation

ICCV 2005 Beijing, Short Course, Oct 15

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. ICCV 2005 Beijing, Short Course, Oct 15 Recognizing and Learning Object Categories Li Fei-Fei, UIUC Rob Fergus, MIT Antonio Torralba, MIT

  2. Agenda • Introduction • Bag of words models • Part-based models • Discriminative methods • Segmentation and recognition • Conclusions

  3. material thing vision perceptible

  4. It's all about programming!

  5. Plato said… • Ordinary objects are classified together if they `participate' in the same abstract Form, such as the Form of a Human or the Form of Quartz. • Forms are proper subjects of philosophical investigation, for they have the highest degree of reality. • Ordinary objects, such as humans, trees, and stones, have a lower degree of reality than the Forms. • Fictions, shadows, and the like have a still lower degree of reality than ordinary objects and so are not proper subjects of philosophical enquiry.

  6. Bruegel, 1564

  7. How many object categories are there? ~10,000 to 30,000 Biederman 1987

  8. So what does object recognition involve?

  9. Verification: is that a bus?

  10. Detection: are there cars?

  11. Identification: is that a picture of Mao?

  12. Object categorization sky building flag face banner wall street lamp bus bus cars

  13. Scene and context categorization • outdoor • city • traffic • …

  14. Challenges 1: view point variation Michelangelo 1475-1564

  15. Challenges 2: illumination slide credit: S. Ullman

  16. Challenges 3: occlusion Magritte, 1957

  17. Challenges 4: scale

  18. Challenges 5: deformation Xu, Beihong 1943

  19. Challenges 6: background clutter Klimt, 1913

  20. History: single object recognition

  21. History: single object recognition • Lowe, et al. 1999, 2003 • Mahamud and Herbert, 2000 • Ferrari, Tuytelaars, and Van Gool, 2004 • Rothganger, Lazebnik, and Ponce, 2004 • Moreels and Perona, 2005 • …

  22. Challenges 7: intra-class variation

  23. History: early object categorization

  24. Turk and Pentland, 1991 • Belhumeur et al. 1997 • Schneiderman et al. 2004 • Viola and Jones, 2000 • Amit and Geman, 1999 • LeCun et al. 1998 • Belongie and Malik, 2002 • Schneiderman et al. 2004 • Argawal and Roth, 2002 • Poggio et al. 1993

  25. ~10,000 to 30,000

  26. OBJECTS INANIMATE ANIMALS PLANTS NATURAL MAN-MADE ….. VERTEBRATE MAMMALS BIRDS TAPIR BOAR GROUSE CAMERA

  27. Scenes, Objects, and Parts Scene Objects Parts Features E. Sudderth, A. Torralba, W. Freeman, A. Willsky. ICCV 2005.

  28. vs. • Bayes rule: posterior ratio likelihood ratio prior ratio Object categorization: the statistical viewpoint

  29. Object categorization: the statistical viewpoint posterior ratio likelihood ratio prior ratio • Discriminative methods model posterior • Generative methods model likelihood and prior

  30. Discriminative • Direct modeling of Decisionboundary Zebra Non-zebra

  31. Generative • Model and

  32. Three main issues • Representation • How to represent an object category • Learning • How to form the classifier, given training data • Recognition • How the classifier is to be used on novel data

  33. Representation • Generative / discriminative / hybrid

  34. Representation • Generative / discriminative / hybrid • Appearance only or location and appearance

  35. Representation • Generative / discriminative / hybrid • Appearance only or location and appearance • Invariances • View point • Illumination • Occlusion • Scale • Deformation • Clutter • etc.

  36. Representation • Generative / discriminative / hybrid • Appearance only or location and appearance • invariances • Part-based or global w/sub-window

  37. Representation • Generative / discriminative / hybrid • Appearance only or location and appearance • invariances • Parts or global w/sub-window • Use set of features or each pixel in image

  38. Learning • Unclear how to model categories, so we learn what distinguishes them rather than manually specify the difference -- hence current interest in machine learning

  39. Learning • Unclear how to model categories, so we learn what distinguishes them rather than manually specify the difference -- hence current interest in machine learning) • Methods of training: generative vs. discriminative

  40. Learning • Unclear how to model categories, so we learn what distinguishes them rather than manually specify the difference -- hence current interest in machine learning) • What are you maximizing? Likelihood (Gen.) or performances on train/validation set (Disc.) • Level of supervision • Manual segmentation; bounding box; image labels; noisy labels Contains a motorbike

  41. Learning • Unclear how to model categories, so we learn what distinguishes them rather than manually specify the difference -- hence current interest in machine learning) • What are you maximizing? Likelihood (Gen.) or performances on train/validation set (Disc.) • Level of supervision • Manual segmentation; bounding box; image labels; noisy labels • Batch/incremental (on category and image level; user-feedback )

  42. Learning • Unclear how to model categories, so we learn what distinguishes them rather than manually specify the difference -- hence current interest in machine learning) • What are you maximizing? Likelihood (Gen.) or performances on train/validation set (Disc.) • Level of supervision • Manual segmentation; bounding box; image labels; noisy labels • Batch/incremental (on category and image level; user-feedback ) • Training images: • Issue of overfitting • Negative images for discriminative methods Priors

  43. Learning • Unclear how to model categories, so we learn what distinguishes them rather than manually specify the difference -- hence current interest in machine learning) • What are you maximizing? Likelihood (Gen.) or performances on train/validation set (Disc.) • Level of supervision • Manual segmentation; bounding box; image labels; noisy labels • Batch/incremental (on category and image level; user-feedback ) • Training images: • Issue of overfitting • Negative images for discriminative methods • Priors

  44. Recognition • Scale / orientation range to search over • Speed

More Related