文件列表:

4Occluded.mat (300001, 2019-01-17)
ComputeSigma.m (1222, 2020-01-30)
ComputeSigmaPart2.m (426, 2020-01-27)
CostElipseGralpolarPDF_3Part2.m (1712, 2020-02-08)
EllipseFittingCA.m (1505, 2020-02-17)
Ellipse_fit.fig (42958, 2020-01-30)
FunctionOptimizaElipse.cpp (10941, 2020-02-04)
FunctionOptimizaElipse.mexmaci64 (17980, 2019-01-17)
FunctionOptimizaElipse.mexw64 (15360, 2020-01-27)
GMMs_Obs拟合椭圆edge map.fig (5520044, 2020-02-04)
GetGaussParam.m (359, 2020-01-28)
Gmms_show.m (783, 2020-01-29)
ImgRelationTest.m (394, 2020-02-04)
LICENSE (1065, 2019-01-17)
Laplas_edge.m (286, 2020-02-04)
Obs_GMMs_俯视.fig (5843579, 2020-02-02)
PlotElipse.m (304, 2020-01-29)
PlotElipse2.m (305, 2020-01-29)
ProcesaImageCA.m (1842, 2020-02-17)
ToOptimizeCirclePart2.m (1448, 2020-02-08)
WZK_ComputeSigmaPart.m (506, 2020-02-07)
Wzk_FitGMMs_Show.m (1413, 2020-02-08)
Wzk_GMMs_Show.m (1162, 2020-02-07)
Wzk_Obs_GMMs.m (373, 2020-02-01)
ellipse6.jpg (3830, 2012-04-26)
ellipse6_GMMs_obs.fig (5896856, 2020-02-04)
ellipse6_L.jpg (12676, 2020-02-04)
ellipse6_R.jpg (12344, 2020-02-04)
ellipse6_small.jpg (16005, 2020-02-04)
ellipse_diy.jpg (18440, 2020-02-07)
ellipse_diy_D.jpg (13923, 2020-02-08)
ellipse_diy_D_L.jpg (14797, 2020-02-08)
ellipse_diy_D_R.jpg (14492, 2020-02-08)
ellipse_diy_L.jpg (14576, 2020-02-08)
ellipse_diy_R.jpg (14598, 2020-02-08)
ellipse_diy_U.jpg (13974, 2020-02-08)
ellipse_diy_obs_GMMs.fig (15209712, 2020-02-07)
normaCA.m (46, 2019-01-17)
... ...

# L2-Ellipse-Fitting This file contains a demo of the paper: Robust Ellipse Detection With Gaussian Mixture Models # Abstract The Euclidian distance between Gaussian Mixtures has been shown to be robust to perform point set registration (Jian and Vemuri, 2011). We propose to extend this idea for robustly matching a family of shapes (ellipses). Optimisation is performed with an annealing strategy, and the search for occurrences is repeated several times to detect multiple instances of the shape of interest. We compare experimentally our approach to other state-of-the-art techniques on a benchmark database for ellipses, and demonstrate the good performance of our approach. # DataSet It uses the 4Occluded.mat benchmark data set. This set contains 50 images with 4 occluded ellipses. It is widelly used for comparing ellipse detection algorithms. # Run Demo To run the algorithm use the following file: [x1,xmo,COSTFinal]=EllipseFittingCA(p) p: Number from 1 to 50 (It corresponds to the image to test in the data set 4Occluded.mat. x1: The resulting ellipse fitted to the data xmo: The starting point (randomly selected) COSTFinal: The final value of the cost function In the paper, this algorithm is used iterativelly to detect all the instances of the ellipse in the image. # Paper reference @article{ARELLANO201612, title = "Robust ellipse detection with Gaussian mixture models", journal = "Pattern Recognition", volume = "58", pages = "12 - 26", year = "2016", issn = "0031-3203", doi = "https://doi.org/10.1016/j.patcog.2016.01.017", url = "http://www.sciencedirect.com/science/article/pii/S0031320316000388", author = "Claudia Arellano and Rozenn Dahyot", keywords = "Ellipse detection, L2 distance, GMM, Parameter estimation", abstract = "The Euclidian distance between Gaussian Mixtures has been shown to be robust to perform point set registration (Jian and Vemuri, 2011). We propose to extend this idea for robustly matching a family of shapes (ellipses). Optimisation is performed with an annealing strategy, and the search for occurrences is repeated several times to detect multiple instances of the shape of interest. We compare experimentally our approach to other state-of-the-art techniques on a benchmark database for ellipses, and demonstrate the good performance of our approach." }

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：