Oriented Edge Forests for Boundary Detection
We present a simple, efﬁcient model for learning bound- ary detection based on a random forest classiﬁer. Our ap- proach combines (1) efﬁcient clustering of training exam- ples based on a simple partitioning of the space of local edge orientations and (2) scale-dependent calibration of in- dividual tree output probabilities prior to multiscale combi- nation. The resulting model outperforms published results on the challenging BSDS500 boundary detection bench- mark. Further, on large datasets our model requires sub- stantially less memory for training and speeds up training time by a factor of 10 over the structured forest model.
Text ReferenceSam Hallman and Charless C. Fowlkes. Oriented edge forests for boundary detection. In CVPR. 2015.
AUTHOR = "Hallman, Sam and Fowlkes, Charless C.",
TITLE = "Oriented Edge Forests for Boundary Detection",
BOOKTITLE = "CVPR",
YEAR = "2015",
tag = "grouping"