Presentation is loading. Please wait.

Presentation is loading. Please wait.

C OMPUTATION M ODEL FOR V ISUAL C ATEGORIZATION Bhuwan Dhingra.

Similar presentations


Presentation on theme: "C OMPUTATION M ODEL FOR V ISUAL C ATEGORIZATION Bhuwan Dhingra."— Presentation transcript:

1 C OMPUTATION M ODEL FOR V ISUAL C ATEGORIZATION Bhuwan Dhingra

2 O VERVIEW Objective: To study the hierarchy of object categorization using a computational model for vision. Three levels of categorization – super-ordinate, basic and subordinate. Basic level categories – maximize cue validities, and dominate any taxonomy. Categorization implemented in unsupervised manner in the current model.

3 H YPOTHESES Rosch et al, [1], claim that basic level categories accessed first. Marc and Joubert, [2], claim that in a purely visual task super-ordinate categories accessed first. Role of expertise emphasized several times in the literature, [3].

4 T HE M ODEL Bag-of-Features:

5 T HE M ODEL Extracted histograms clustered in an unsupervised manner using k-means algorithm. Distance metric used – (1-correlation(h1,h2)), where h1 and h2 are two histograms.

6 D ATASET 30 images for each subordinate category using Google image search of the keywords.

7 D ATASET Furniture Animal TableChairBirdDog Coffee Table Picnic Table Rocking Chair Bar- stool Crow Pigeon Foxhound Dalmation Super-ordinate classes Basic classes Sub-ordinate classes

8 T ESTS Test 1: Study which type of categorization dominates as the number of detected key-points is varied. Test 2: Study how the performance of the categorization changes with the number of images. Test 3: Study the effect of increasing the number of images of one basic category compared to others Different categorizations were implemented by setting k = 2,4,8.

9 P ERFORMANCE I NDICES Rand Index: TP, TN, FP, FN are true positive and negatives, and false positives and negatives. Purity: Percentage of correctly assigned points, assuming majority class for each cluster. Normalized Mutual Information: Information theoretic mutual information between clusters and classes (normalized to 1). Silhouette Index: Based on the ratio of the within class scatter to between class scatter.

10 R ESULTS Variation of the performance metrics with Peak Threshold or the number of key-points detected.

11 R ESULTS Variation of the performance metrics with Peak Threshold or the number of key-points detected.

12 R ESULTS Variation of the performance metrics with Peak Threshold or the number of key-points detected.

13 R ESULTS Variation of performance metrics with number of images:

14 R ESULTS Effect of expertise Two subordinate and one basic level categories taken together, ex: {{dalmation, foxhound}, bird} Trial 1: Training samples of subordinate categories half of basic category Trial 2: Training samples of subordinate category equal to basic category

15 S OME P ROBLEMS White background images sometimes classified separate from cluttered background. Solution: Foreground extraction High variability in Normalized Mutual Information (NMI) Effect of expertise not clear Solution: Test for exponential increase in images

16 R EFERENCES [1] Rosch, E., Mervis, C., Gray, W., Johnson, D., & Boyes-Braem, P. (1976). Basic objects in natural categories. Cognitive Psychology.Basic objects in natural categories [2] Marc, J.M.M., Joubert, O.R., Nespoulous, J.L. & Fabre-Thorpe, M (2009). The time-course of visual categorizations: you spot the animal faster than the bird. PLoS one.The time-course of visual categorizations: you spot the animal faster than the bird [3] Johnson, K.E., Mervis, C.B. (1997). Effects of varying levels of expertise on the basic level of categorization. Journal of Expert Psychology.Effects of varying levels of expertise on the basic level of categorization


Download ppt "C OMPUTATION M ODEL FOR V ISUAL C ATEGORIZATION Bhuwan Dhingra."

Similar presentations


Ads by Google