Back to Algorithm Ranking.

[Color] Boundary Detection Benchmark: Algorithm "Brightness / Color / Texture Gradients"

This algorithm uses a combination of local brightness, color, and texture gradients to detect boundaries. The gradients are combined with a logistic to model the posterior probability of a boundary. Each gradient is computed as the chi-squared difference in the distribution of some feature in two half discs centered at a pixel and divided in half at the putative boundary orientation. Using the 1976 CIELAB colorspace, the brightness and color gradients are given by the difference in luminance and chrominance distributions. The texture gradient is given by the difference in texton distributions. Textons are computed by clustering filterbank responses using k-means, so that they model the joint distribution of filter responses. The filters are standard even- and odd-symmetric quadrature pair elongated linear filters. 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=0.79
#2 (170057) F=0.59
#3 (58060) F=0.64
#4 (163085) F=0.51
#5 (42049) F=0.86
#6 (167062) F=0.88
#7 (157055) F=0.76
#8 (295087) F=0.71
#9 (24077) F=0.68
#10 (78004) F=0.78
#11 (220075) F=0.61
#12 (45096) F=0.77
#13 (38092) F=0.74
#14 (43074) F=0.69
#15 (16077) F=0.59
#16 (86000) F=0.69
#17 (101085) F=0.79
#18 (219090) F=0.67
#19 (89072) F=0.57
#20 (300091) F=0.78
#21 (126007) F=0.74
#22 (156065) F=0.65
#23 (76053) F=0.65
#24 (296007) F=0.71
#25 (175032) F=0.61
#26 (253027) F=0.73
#27 (304034) F=0.44
#28 (86016) F=0.59
#29 (103070) F=0.65
#30 (8023) F=0.36
#31 (260058) F=0.68
#32 (41033) F=0.68
#33 (291000) F=0.68
#34 (109053) F=0.53
#35 (130026) F=0.48
#36 (241004) F=0.79
#37 (108082) F=0.35
#38 (285079) F=0.75
#39 (147091) F=0.76
#40 (69040) F=0.46
#41 (14037) F=0.66
#42 (54082) F=0.68
#43 (189080) F=0.75
#44 (229036) F=0.79
#45 (62096) F=0.82
#46 (271035) F=0.69
#47 (167083) F=0.74
#48 (12084) F=0.42
#49 (69015) F=0.79
#50 (148089) F=0.57
#51 (160068) F=0.59
#52 (145086) F=0.82
#53 (216081) F=0.81
#54 (97033) F=0.72
#55 (182053) F=0.73
#56 (208001) F=0.65
#57 (19021) F=0.70
#58 (227092) F=0.83
#59 (134035) F=0.75
#60 (223061) F=0.70
#61 (253055) F=0.63
#62 (148026) F=0.52
#63 (210088) F=0.54
#64 (86068) F=0.58
#65 (3096) F=0.75
#66 (41069) F=0.71
#67 (21077) F=0.74
#68 (196073) F=0.84
#69 (108070) F=0.40
#70 (123074) F=0.50
#71 (376043) F=0.68
#72 (306005) F=0.69
#73 (38082) F=0.60
#74 (33039) F=0.62
#75 (108005) F=0.56
#76 (106024) F=0.68
#77 (302008) F=0.68
#78 (102061) F=0.51
#79 (197017) F=0.80
#80 (299086) F=0.80
#81 (37073) F=0.83
#82 (241048) F=0.76
#83 (65033) F=0.71
#84 (55073) F=0.62
#85 (66053) F=0.71
#86 (143090) F=0.69
#87 (85048) F=0.75
#88 (42012) F=0.61
#89 (351093) F=0.73
#90 (361010) F=0.76
#91 (175043) F=0.75
#92 (87046) F=0.65
#93 (105025) F=0.64
#94 (236037) F=0.54
#95 (101087) F=0.78
#96 (304074) F=0.65
#97 (296059) F=0.84
#98 (159008) F=0.63
#99 (385039) F=0.75
#100 (69020) F=0.72

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