1. Object Recognition

2. So what does object recognition involve?

3. Verification: is that a bus?

4. Detection: are there cars?

5. Identification: is that a picture of Mao?

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

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

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

9. Challenges 3: occlusion Magritte, 1957

10. Challenges 4: scale

11. Challenges 5: deformation Xu, Beihong 1943

12. Challenges 7: intra-class variation

13. Two main approaches Part-based Global sub-window

14. Global Approaches x1x1 x2x2 x3x3 Vectors in high- dimensional space Aligned images

15. x1x1 x2x2 x3x3 Vectors in high-dimensional space Global Approaches Training Involves some dimensionality reduction Detector

16. –Scale / position range to search over Detection

17. –Scale / position range to search over

18. Detection –Scale / position range to search over

19. Detection –Combine detection over space and scale.

20. PROJECT 1 Build a detection system that inputs an image, runs a detector over (x,y) and scales, and removes spurious detections. The system should be able to run different detectors. For initial testing use linear SVM (existing package). Challenge: Algorithm for integration of raw detections. Speed.

21. Turk and Pentland, 1991 Belhumeur et al. 1997 Schneiderman et al. 2004 Viola and Jones, 2000 Keren et al. 2001 Osadchy et al. 2004 Amit and Geman, 1999 LeCun et al. 1998 Belongie and Malik, 2002 Schneiderman et al. 2004 Argawal and Roth, 2002 Poggio et al. 1993

22. Antiface method for detection No training on negative examples is required. A set of rejectors is applied in cascaded manner. Robust to large pose variation. Simple and very fast.

23. Intuition Lower probability image smoothness measure Boltzmann distribution How are the natural images distributed in a high dimensional space?

24. Lower probability Antiface Much less false positives PCA Many false positives Intuition

25. Main Idea Claim: for random natural images viewed as unit vectors, is large on average. – for all positive class – d is smooth is large on average for random natural image. Anti-Face detector is defined as a vector d satisfying:

26. Discrimination SMALL LARGE If x is an image and  is a target class:

27. Cascade of Independent Detectors 7 inner products 4 inner products

28. Example Samples from the training set 4 Anti-Face Detectors

29. 4 Anti-face Detectors

30. Eigenface method with the subspace of dimension 100

31. PROJECT 2 Implement Antiface method for detection*. Implement several extensions of Antifaces: –Change the accepting rule so that instead of passing all the detectors it passes at least 80% of detectors. – Apply Naïve Bayes in 10D antiface space –Project each image onto 20D Antiface space and train SVM in this space. See project page for details * D. Keren M. Osadchy and C. Gotsman, Anti-Faces: A novel, fast method for image detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, No. 7, July 2001, pp. 747-761.

32. Part-Based Approaches Object Bag of ‘words’ Constellation of parts

33. Of all the sensory impressions proceeding to the brain, the visual experiences are the dominant ones. Our perception of the world around us is based essentially on the messages that reach the brain from our eyes. For a long time it was thought that the retinal image was transmitted point by point to visual centers in the brain; the cerebral cortex was a movie screen, so to speak, upon which the image in the eye was projected. Through the discoveries of Hubel and Wiesel we now know that behind the origin of the visual perception in the brain there is a considerably more complicated course of events. By following the visual impulses along their path to the various cell layers of the optical cortex, Hubel and Wiesel have been able to demonstrate that the message about the image falling on the retina undergoes a step- wise analysis in a system of nerve cells stored in columns. In this system each cell has its specific function and is responsible for a specific detail in the pattern of the retinal image. sensory, brain, visual, perception, retinal, cerebral cortex, eye, cell, optical nerve, image Hubel, Wiesel China is forecasting a trade surplus of $90bn (£51bn) to $100bn this year, a threefold increase on 2004's $32bn. The Commerce Ministry said the surplus would be created by a predicted 30% jump in exports to $750bn, compared with a 18% rise in imports to $660bn. The figures are likely to further annoy the US, which has long argued that China's exports are unfairly helped by a deliberately undervalued yuan. Beijing agrees the surplus is too high, but says the yuan is only one factor. Bank of China governor Zhou Xiaochuan said the country also needed to do more to boost domestic demand so more goods stayed within the country. China increased the value of the yuan against the dollar by 2.1% in July and permitted it to trade within a narrow band, but the US wants the yuan to be allowed to trade freely. However, Beijing has made it clear that it will take its time and tread carefully before allowing the yuan to rise further in value. China, trade, surplus, commerce, exports, imports, US, yuan, bank, domestic, foreign, increase, trade, value Bag of ‘words’ analogy to documents

35. Interest Point Detectors Basic requirements: –Sparse –Informative –Repeatable Invariance –Rotation –Scale (Similarity) –Affine

36. Popular Detectors Scale Invariant Affine Invariant Harris-Laplace Affine Difference of GaussiansLaplace of GaussiansScale Saliency (Kadir- Braidy) Harris-Laplace Difference of Gaussians Affine Laplace of Gaussians Affine Affine Saliency (Kadir- Braidy) The are many others… See: 1)“Scale and affine invariant interest point detectors” K. Mikolajczyk, C. Schmid, IJCV, Volume 60, Number 1 - 2004 2)“A comparison of affine region detectors”, K. Mikolajczyk, T. Tuytelaars, C. Schmid, A. Zisserman, J. Matas, F. Schaffalitzky, T. Kadir and L. Van Gool, http://www.robots.ox.ac.uk/~vgg/research/affine/det_eval_files/vibes_ijcv2004.pdf http://www.robots.ox.ac.uk/~vgg/research/affine/det_eval_files/vibes_ijcv2004.pdf

37. Representation of appearance: Local Descriptors Invariance –Rotation –Scale –Affine Insensitive to small deformations Illumination invariance –Normalize out

38. SIFT – Scale Invariant Feature Transform Descriptor overview: –Determine scale (by maximizing DoG in scale and in space), local orientation as the dominant gradient direction. Use this scale and orientation to make all further computations invariant to scale and rotation. –Compute gradient orientation histograms of several small windows (128 values for each point) –Normalize the descriptor to make it invariant to intensity change David G. Lowe, "Distinctive image features from scale-invariant keypoints,“ International Journal of Computer Vision, 60, 2 (2004), pp. 91-110.

39. Feature Detection and Representation Normalize patch Detect patches [Mikojaczyk and Schmid ’02] [Matas et al. ’02] [Sivic et al. ’03] Compute SIFT descriptor [Lowe’99] Slide credit: Josef Sivic

40. … Feature Detection and Representation

41. Codewords dictionary formation …

42. Vector quantization … Slide credit: Josef Sivic

43. Codewords dictionary formation Fei-Fei et al. 2005

44. Image patch examples of codewords Sivic et al. 2005

45. Vector X Representation Learning positive negative SVM classifier positive negative SVM classification

46. Recognition SVM(X) Contains object Vector X Representation Doesn’t contain object

47. PROJECT 3 Implement a bag of ‘words’ approach. The method is described in “Visual Categorization with Bags of Keypoints” G.Cruska, C. R. Dance, L.Fan, J.Willamowski,C. Bray. Test it on 4 categories (from 101 database): airplanes, faces, cars side, motorbikes, against background.

48. PROJECT 4 Implement part based method, described in “Class Recognition Using Discriminative Local Features”, by G. Dorkó, C. Schmid. Test it on Oxford object data set. Compare the performance of the algorithm using different point detectors. The code for point detectors is provided. Compare the performance of the algorithm with original SIFT and with SIFT without rotation invariance. The initial code for SIFT is provided, but should be edited to remove rotation invariance.