资 源 简 介
This paper investigates two fundamental problems in computer vision: contour detection and image segmentation. We
present state-of-the-art algorithms for both of these tasks. Our contour detector combines multiple local cues into a globalization
framework based on spectral clustering. Our segmentation algorithm consists of generic machinery for transforming the output of
any contour detector into a hierarchical region tree. In this manner, we reduce the problem of image segmentation to that of contour
detection. Extensive experimental evaluation demonstrates that both our contour detection and segmentation methods significantly
outperform competing algorithms. The automatically generated hierarchical segmentations can be interactively refined by user-
specified annotations. Computation at multiple image resolutions provides a means of coupling our system to recognition applications.
文 件 列 表
lib
applyFilter.m
buildW.mexa64
buildW.mexmaci64
contours2ucm.m
det_mPb.m
det_mPb_lg.m
det_mPb_sm.m
fbRun.m
fitparab.m
fit_contour.m
globalPb.m
globalPb_im.m
globalPb_im_multiscale.m
globalPb_pieces.m
gPb_from_cues.m
im2ucm.m
isum.m
load_smatrix.mexa64
load_smatrix.mexmaci64
mex_contour_sides.mexa64
mex_contour_sides.mexmaci64
mex_line_inds.mexa64
mex_line_inds.mexmaci64
mex_nonmax_oriented.mexa64
mex_nonmax_oriented.mexmaci64
mex_pb_parts_final_selected.mexa64
mex_pb_parts_final_selected.mexmaci64
mex_pb_parts_lg.mexa64
mex_pb_parts_lg.mexmaci64
mex_pb_parts_sm.mexa64
mex_pb_parts_sm.mexmaci64
mPb_from_cues.m
multiscalePb.m
nonmax_channels.m
nonmax_oriented.m
oeFilter.m
padReflect.m
savgol.m
savgol_border.mexa64
savgol_border.mexmaci64
spectralPb.m
ucm_mean_pb.mexa64
ucm_mean_pb.mexmaci64
uvt.mexa64
uvt.mexmaci64
interactive
LICENSE
AGPL
101087_seeds.bmp
example_interactive.m
interactive_segmentation.m
data
101087.jpg
101087_big.jpg
101087_gPb.mat
101087_small.jpg
101087_ucm.bmp
101087_ucm2.mat
example.m
run_bsds500.m
test_gPb_ucm.m
Arbelaez_Maire_Fowlkes_Malik_TPAMI2010.pdf