[Color] Boundary Detection Benchmark: Algorithm "Color Gradient"

This algorithm uses local color gradients to detect boundaries. The A and B gradient channels (from CIELAB) are combined with a logistic to model the posterior probability of a boundary. The color gradient is computed as the chi-squared difference in the distribution of pixel chrominance values in two half discs centered at a pixel and divided in half at the putative boundary orientation. See our NIPS and PAMI papers in the Grouping area of the Berkeley Vision Group web pages for additional details.

Click on an image for additional details.

#1 (119082) F=0.62

#2 (170057) F=0.55

#3 (58060) F=0.61

#4 (163085) F=0.43

#5 (42049) F=0.75

#6 (167062) F=0.45

#7 (157055) F=0.63

#8 (295087) F=0.68

#9 (24077) F=0.54

#10 (78004) F=0.62

#11 (220075) F=0.57

#12 (45096) F=0.74

#13 (38092) F=0.63

#14 (43074) F=0.54

#15 (16077) F=0.48

#16 (86000) F=0.64

#17 (101085) F=0.81

#18 (219090) F=0.65

#19 (89072) F=0.53

#20 (300091) F=0.69

#21 (126007) F=0.68

#22 (156065) F=0.62

#23 (76053) F=0.61

#24 (296007) F=0.67

#25 (175032) F=0.52

#26 (253027) F=0.59

#27 (304034) F=0.57

#28 (86016) F=0.47

#29 (103070) F=0.55

#30 (8023) F=0.35

#31 (260058) F=0.68

#32 (41033) F=0.46

#33 (291000) F=0.65

#34 (109053) F=0.51

#35 (130026) F=0.43

#36 (241004) F=0.77

#37 (108082) F=0.32

#38 (285079) F=0.65

#39 (147091) F=0.46

#40 (69040) F=0.44

#41 (14037) F=0.62

#42 (54082) F=0.56

#43 (189080) F=0.70

#44 (229036) F=0.72

#45 (62096) F=0.59

#46 (271035) F=0.69

#47 (167083) F=0.77

#48 (12084) F=0.34

#49 (69015) F=0.65

#50 (148089) F=0.55

#51 (160068) F=0.50

#52 (145086) F=0.63

#53 (216081) F=0.74

#54 (97033) F=0.57

#55 (182053) F=0.59

#56 (208001) F=0.63

#57 (19021) F=0.60

#58 (227092) F=0.56

#59 (134035) F=0.65

#60 (223061) F=0.58

#61 (253055) F=0.52

#62 (148026) F=0.61

#63 (210088) F=0.43

#64 (86068) F=0.46

#65 (3096) F=0.68

#66 (41069) F=0.54

#67 (21077) F=0.57

#68 (196073) F=0.59

#69 (108070) F=0.33

#70 (123074) F=0.43

#71 (376043) F=0.66

#72 (306005) F=0.56

#73 (38082) F=0.64

#74 (33039) F=0.50

#75 (108005) F=0.54

#76 (106024) F=0.48

#77 (302008) F=0.60

#78 (102061) F=0.56

#79 (197017) F=0.54

#80 (299086) F=0.68

#81 (37073) F=0.67

#82 (241048) F=0.72

#83 (65033) F=0.59

#84 (55073) F=0.57

#85 (66053) F=0.66

#86 (143090) F=0.59

#87 (85048) F=0.60

#88 (42012) F=0.55

#89 (351093) F=0.57

#90 (361010) F=0.68

#91 (175043) F=0.66

#92 (87046) F=0.49

#93 (105025) F=0.46

#94 (236037) F=0.53

#95 (101087) F=0.68

#96 (304074) F=0.57

#97 (296059) F=0.78

#98 (159008) F=0.63

#99 (385039) F=0.73

#100 (69020) F=0.77

Page generated on 20-Feb-2013 11:08:23.