rpnp1.0
所属分类：matlab编程
开发工具：matlab
文件大小：83KB
下载次数：59
上传日期：2015-08-14 16:31:45
上 传 者：
moyanlane
说明： 姿态估计，包好3种方法，其中有经典的pnp,rpnp，hlm等等，以及他们的比较。
(Pose estimation, package 3 methods, including the classic PNP, rpnp, HLM, and so on, and their comparison.)
文件列表:
rpnp1.0\code3\epnp\compute_alphas.m (1050, 2007-09-05)
rpnp1.0\code3\epnp\compute_A_and_b_Gauss_Newton.m (1522, 2007-10-27)
rpnp1.0\code3\epnp\compute_constraint_distance_2param_6eq_3unk.m (4130, 2007-09-05)
rpnp1.0\code3\epnp\compute_constraint_distance_3param_6eq_6unk.m (7573, 2007-09-05)
rpnp1.0\code3\epnp\compute_constraint_distance_orthog_4param_9eq_10unk.m (21501, 2007-09-05)
rpnp1.0\code3\epnp\compute_L6_10.m (6792, 2007-10-27)
rpnp1.0\code3\epnp\compute_M_ver2.m (1466, 2007-09-05)
rpnp1.0\code3\epnp\compute_norm_sign_scaling_factor.m (1969, 2007-10-27)
rpnp1.0\code3\epnp\compute_permutation_constraint4.m (2639, 2007-09-05)
rpnp1.0\code3\epnp\compute_rho.m (1044, 2007-10-27)
rpnp1.0\code3\epnp\define_control_points.m (880, 2007-09-05)
rpnp1.0\code3\epnp\define_distances_btw_control_points.m (1328, 2007-09-05)
rpnp1.0\code3\epnp\dist2.m (865, 2007-10-27)
rpnp1.0\code3\epnp\efficient_pnp.m (6775, 2009-11-02)
rpnp1.0\code3\epnp\efficient_pnp_gauss.m (8268, 2007-10-27)
rpnp1.0\code3\epnp\efficient_pnp_planar.m (16913, 2011-08-02)
rpnp1.0\code3\epnp\gauss_newton.m (1465, 2007-10-27)
rpnp1.0\code3\epnp\getrotT.m (537, 2007-09-05)
rpnp1.0\code3\epnp\kernel_noise.m (878, 2007-09-05)
rpnp1.0\code3\epnp\optimize_betas_gauss_newton.m (1733, 2007-10-27)
rpnp1.0\code3\epnp\reprojection_error_usingRT.m (1170, 2009-08-16)
rpnp1.0\code3\epnp\sign_determinant.m (985, 2007-10-27)
rpnp1.0\code3\func\cal_pose_err.m (345, 2010-07-28)
rpnp1.0\code3\func\DLT.m (927, 2010-10-30)
rpnp1.0\code3\func\EPnP.m (88, 2010-07-26)
rpnp1.0\code3\func\EPnP_GN.m (97, 2010-07-26)
rpnp1.0\code3\func\HOMO.m (512, 2010-07-28)
rpnp1.0\code3\func\LHM.m (95, 2010-07-26)
rpnp1.0\code3\func\rodrigues.m (4938, 2008-04-15)
rpnp1.0\code3\func\RPnP.m (6556, 2011-08-12)
rpnp1.0\code3\func\RPnP1.m (6477, 2011-08-12)
rpnp1.0\code3\func\RPnP2.m (6526, 2011-08-02)
rpnp1.0\code3\func\showpercent.m (241, 2009-05-08)
rpnp1.0\code3\func\SP.m (294, 2010-07-28)
rpnp1.0\code3\func\xdrawgraph.m (531, 2010-07-27)
rpnp1.0\code3\func\xrand.m (68, 2009-09-12)
rpnp1.0\code3\lhm\objpose.m (6619, 2010-04-24)
rpnp1.0\code3\lhm\qmatQ.m (195, 2005-06-06)
rpnp1.0\code3\lhm\qmatW.m (196, 2005-06-06)
rpnp1.0\code3\lhm\quat2mat.m (284, 2005-06-06)
... ...
==================================
Please run the following files in matlab (recommended version: R2008a)
(1) main_ordinary_3d.m
% The experiment in the ordinary 3D case.
% noise level 3 pixels.
% number of points n= 4 ... 20
% The methods compared are DLT, EPnP, EPnP_GN, LHM and RPnP.
% Details can be found in the paper.
(2) main_quasi_singular.m
% The experiment in the quasi-singular case.
% noise level 3 pixels.
% number of points n= 4 ... 20
% The methods compared are DLT, EPnP, EPnP_GN, LHM and RPnP.
(3) main_planar.m
% The experiment in the planar case.
% noise level 3 pixels.
% number of points n= 4 ... 20
% The methods compared are HOMO, LHM, SP and RPnP.
% The EPnP code for planar target is not involved, because it has not been publicly released.
(4) main_selecting_axis.m
% The experiment on the influence of the rotation axis selection step.
% The rotation axis is selected in different ways:
% RPnP1: randomly select an edge
% RPnP*: finding the longest in all n(n-1)/2 edges
% RPnP : the default setting which randomly semples n edge and selects the longest
(5) main_time.m
% The experiment which tests the computational time.
% LHM, EPnP, EPnP_GN and RPnP are tested.
(6) main_4pt.m
% The experiment in the ordinary 3D case.
% noise level 0.5 ... 5 pixels.
% number of points n= 4
% The methods compared are DLT, EPnP, EPnP_GN, LHM and RPnP.
(7) main_5pt.m
% The experiment in the ordinary 3D case.
% noise level 0.5 ... 5 pixels.
% number of points n= 5
% The methods compared are DLT, EPnP, EPnP_GN, LHM and RPnP.
=================================
The method list:
(1) ./func/RPnP.m
% The RPnP algorithm by C. Xu
% function [R t]= RPnP(XX,xx)
% XX is the 3D coordinate of the point set.
% xx is the ***normalized*** 2D coordinate of the projected point set.
% R is the estimated rotation matrix.
% t is the estimated translation vector.
(2) ./epnp/
% The EPnP algorithm by F. Moreno-Noguer, V. Lepetit, P. Fua.
(3) ./lhm/
% The LHM algorithm by Chien-Ping Lu et. al.
(4) ./sp/
% The SP algorithm for planar targets by G. Schweighofer and A. Pinz.
近期下载者：
相关文件：
收藏者：