Hello

First Meeting Resume (Hendro Subagyo)

Need for IR:

• Information Search
• Information Routing
• Information Filterring
• User’s information need

Source from:

http://www.eecs.berkeley.edu/Research/Projects/CS/vision/bsds/bench/html/algorithms.html

Boundary Detection Benchmark: Algorithm Ranking

Summary Tables

Grayscale	Color
Rank Score Algorithm 0 0.79 Humans 1 0.68 Global Probability of Boundary (gray) 2 0.66 xren (gray) 3 0.64 Boosted Edge Learning 4 0.63 Brightness / Texture Gradients 5 0.60 Brightness Gradient 6 0.58 Texture Gradient 7 0.58 Multiscale Gradient Magnitude 8 0.57 Second Moment Matrix 9 0.56 Gradient Magnitude 10 0.48 Segmentation Induced by Scale Invariance 11 0.41 Random	Rank Score Algorithm 0 0.79 Humans 1 0.70 Global Probability of Boundary (color) 2 0.68 xren (color) 3 0.67 Ultrametric Contour Maps 4 0.66 Boosted Edge Learning (color) 5 0.65 min-cover 6 0.65 Brightness / Color / Texture Gradients 7 0.63 Brightness / Texture Gradients 8 0.57 Color Gradient 9 0.43 Random

backlink: jaringankantor.com