[Color] Boundary Detection Benchmark: Algorithm "Ultrametric Contour Maps"

Boundary Extraction in Natural Images Using Ultrametric Contour Maps. [pdf]. POCV 2006.
Pablo Arbelaez.

This algorithm relies on the formulation of the segmentation problem in the framework of hierarchical data classification, in which the geometric structure of an image can be represented by an Ultrametric Contour Map (UCM), the soft boundary image associated to a family of nested segmentations. A threshold at level k on a UCM provides always a set of closed curves, the boundaries of the segmentation at scale k. We define generic ultrametric distances for boundary extraction by integrating local contour cues along the regions' boundaries and combining this information with intra-region attributes.

Click on an image for additional details.

#1 (119082) F=0.83

#2 (170057) F=0.66

#3 (58060) F=0.67

#4 (163085) F=0.51

#5 (42049) F=0.86

#6 (167062) F=0.92

#7 (157055) F=0.80

#8 (295087) F=0.77

#9 (24077) F=0.67

#10 (78004) F=0.81

#11 (220075) F=0.64

#12 (45096) F=0.77

#13 (38092) F=0.77

#14 (43074) F=0.68

#15 (16077) F=0.60

#16 (86000) F=0.69

#17 (101085) F=0.85

#18 (219090) F=0.73

#19 (89072) F=0.58

#20 (300091) F=0.82

#21 (126007) F=0.74

#22 (156065) F=0.78

#23 (76053) F=0.65

#24 (296007) F=0.80

#25 (175032) F=0.65

#26 (253027) F=0.77

#27 (304034) F=0.77

#28 (86016) F=0.57

#29 (103070) F=0.70

#30 (8023) F=0.37

#31 (260058) F=0.76

#32 (41033) F=0.74

#33 (291000) F=0.65

#34 (109053) F=0.56

#35 (130026) F=0.47

#36 (241004) F=0.85

#37 (108082) F=0.41

#38 (285079) F=0.77

#39 (147091) F=0.69

#40 (69040) F=0.49

#41 (14037) F=0.70

#42 (54082) F=0.71

#43 (189080) F=0.77

#44 (229036) F=0.85

#45 (62096) F=0.87

#46 (271035) F=0.72

#47 (167083) F=0.80

#48 (12084) F=0.37

#49 (69015) F=0.82

#50 (148089) F=0.61

#51 (160068) F=0.65

#52 (145086) F=0.80

#53 (216081) F=0.88

#54 (97033) F=0.77

#55 (182053) F=0.79

#56 (208001) F=0.73

#57 (19021) F=0.74

#58 (227092) F=0.83

#59 (134035) F=0.75

#60 (223061) F=0.71

#61 (253055) F=0.72

#62 (148026) F=0.74

#63 (210088) F=0.60

#64 (86068) F=0.62

#65 (3096) F=0.72

#66 (41069) F=0.79

#67 (21077) F=0.76

#68 (196073) F=0.82

#69 (108070) F=0.40

#70 (123074) F=0.59

#71 (376043) F=0.78

#72 (306005) F=0.76

#73 (38082) F=0.65

#74 (33039) F=0.69

#75 (108005) F=0.60

#76 (106024) F=0.76

#77 (302008) F=0.68

#78 (102061) F=0.59

#79 (197017) F=0.83

#80 (299086) F=0.77

#81 (37073) F=0.81

#82 (241048) F=0.78

#83 (65033) F=0.75

#84 (55073) F=0.61

#85 (66053) F=0.77

#86 (143090) F=0.72

#87 (85048) F=0.75

#88 (42012) F=0.64

#89 (351093) F=0.72

#90 (361010) F=0.76

#91 (175043) F=0.75

#92 (87046) F=0.62

#93 (105025) F=0.70

#94 (236037) F=0.57

#95 (101087) F=0.80

#96 (304074) F=0.69

#97 (296059) F=0.86

#98 (159008) F=0.64

#99 (385039) F=0.81

#100 (69020) F=0.84

Page generated on 14-Jan-2009 15:58:02.