Presentation is loading. Please wait.

Presentation is loading. Please wait.

Mimicking human texture classification Eva M. van Rikxoort Egon L. van den Broek Theo E. Schouten.

Similar presentations


Presentation on theme: "Mimicking human texture classification Eva M. van Rikxoort Egon L. van den Broek Theo E. Schouten."— Presentation transcript:

1 Mimicking human texture classification Eva M. van Rikxoort Egon L. van den Broek Theo E. Schouten

2 Mimicking human texture classification Introduction Human texture perception important for CBIR –Humans judge results Mimick by a computer unconstrained human texture clustering –color and gray texture images

3 Mimicking human texture classification Human texture clustering Card Sorting: Visual Short Term Memory: 4 – 14 items Experts: the number of clusters is 6

4 Mimicking human texture classification 180 images, VisTex and OuTex 15 categories

5 Mimicking human texture classification Participants 18 subjects with normal vision –no instruction on what features to use –color and gray reported afterwards –“garbage group” –“gray more difficult”

6 Mimicking human texture classification Results: human texture clustering Consensus matrix per pair of participants Average consensus between all participants: –color: 57% gray: 56% Most prototypical (mp) clusters –core images on which 45% of pairs agree –color: 70% gray: 65% Base images for each mp cluster: –assigned to it by at least 8 participants

7 Mimicking human texture classification All base images for one cluster

8 Mimicking human texture classification Automatic texture clustering K-means clustering algorithm Feature vectors (previous research) –gray 32 bin histogram –4 texture features from co-occurrence matrix –11 color categories histogram –4 texture features from color correlogram

9 Mimicking human texture classification Automatic vs. human texture clustering Consensus matrix pair Overall consensus for each feature vector –Color: 45% - 46% - 47% –Gray: 44% - 45% - 42% each mp cluster (base images) –clusters no automatic images (color 1, gray 2) –clusters which match well –higher agreement for gray –binary and fuzzy measure

10 Mimicking human texture classification Humans judging automatic clustering By giving a mark ranging from 1 to 10 to the homogeneity of an automatic cluster. using the best performing feature vector for color and gray Results (36 participants): –Average rating gray: 6.1, SD 3.1 –Average rating color: 5.2, SD 3.1

11 Mimicking human texture classification The benchmark: a gray-scale cluster

12 Mimicking human texture classification Summary and conclusions little consensus between participants –no generic human texture clustering –multiple strategies in mimicking same overall consensus classifier-human –humans use more semantics in color images –human color categories in classifier gray classifier better on individual clusters –less or better separated semantics by humans

13 Mimicking human texture classification and further more consensus on 50% core images –common part in human strategies “garbage cluster” approach uniform –try to mimic that in classifier

14 The End


Download ppt "Mimicking human texture classification Eva M. van Rikxoort Egon L. van den Broek Theo E. Schouten."

Similar presentations


Ads by Google