[37] | 1 | % * This code was used in the following articles:
|
---|
| 2 | % * [1] Learning 3-D Scene Structure from a Single Still Image,
|
---|
| 3 | % * Ashutosh Saxena, Min Sun, Andrew Y. Ng,
|
---|
| 4 | % * In ICCV workshop on 3D Representation for Recognition (3dRR-07), 2007.
|
---|
| 5 | % * (best paper)
|
---|
| 6 | % * [2] 3-D Reconstruction from Sparse Views using Monocular Vision,
|
---|
| 7 | % * Ashutosh Saxena, Min Sun, Andrew Y. Ng,
|
---|
| 8 | % * In ICCV workshop on Virtual Representations and Modeling
|
---|
| 9 | % * of Large-scale environments (VRML), 2007.
|
---|
| 10 | % * [3] 3-D Depth Reconstruction from a Single Still Image,
|
---|
| 11 | % * Ashutosh Saxena, Sung H. Chung, Andrew Y. Ng.
|
---|
| 12 | % * International Journal of Computer Vision (IJCV), Aug 2007.
|
---|
| 13 | % * [6] Learning Depth from Single Monocular Images,
|
---|
| 14 | % * Ashutosh Saxena, Sung H. Chung, Andrew Y. Ng.
|
---|
| 15 | % * In Neural Information Processing Systems (NIPS) 18, 2005.
|
---|
| 16 | % *
|
---|
| 17 | % * These articles are available at:
|
---|
| 18 | % * http://make3d.stanford.edu/publications
|
---|
| 19 | % *
|
---|
| 20 | % * We request that you cite the papers [1], [3] and [6] in any of
|
---|
| 21 | % * your reports that uses this code.
|
---|
| 22 | % * Further, if you use the code in image3dstiching/ (multiple image version),
|
---|
| 23 | % * then please cite [2].
|
---|
| 24 | % *
|
---|
| 25 | % * If you use the code in third_party/, then PLEASE CITE and follow the
|
---|
| 26 | % * LICENSE OF THE CORRESPONDING THIRD PARTY CODE.
|
---|
| 27 | % *
|
---|
| 28 | % * Finally, this code is for non-commercial use only. For further
|
---|
| 29 | % * information and to obtain a copy of the license, see
|
---|
| 30 | % *
|
---|
| 31 | % * http://make3d.stanford.edu/publications/code
|
---|
| 32 | % *
|
---|
| 33 | % * Also, the software distributed under the License is distributed on an
|
---|
| 34 | % * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either
|
---|
| 35 | % * express or implied. See the License for the specific language governing
|
---|
| 36 | % * permissions and limitations under the License.
|
---|
| 37 | % *
|
---|
| 38 | % */
|
---|
| 39 | function [Pair ImgInfo matches fail]=PoseMatchEst(defaultPara, ImgInfo) |
---|
| 40 | |
---|
| 41 | % This function estimate the relative Pose of the camera using first camera coordinate |
---|
| 42 | % as world coordinate |
---|
| 43 | |
---|
| 44 | % Input: |
---|
| 45 | % default - camera intrinsic, etc |
---|
| 46 | % ImgInfo - Exif, Model info, GPS, IMU info |
---|
| 47 | % |
---|
| 48 | % Return: |
---|
| 49 | % R - rotation - (R*Posi2+ T to A's coordinate) |
---|
| 50 | % T - translation |
---|
| 51 | |
---|
| 52 | % step outline |
---|
| 53 | % 1) extract Measuesd Position and orientation from GPS or IMU info |
---|
| 54 | % 2) Using Measures R and T and Mono-Depth to define mach search space constrain |
---|
| 55 | % 3) Do match search with all combinations satisfying Constrain from 2) using ralative threshould |
---|
| 56 | % 4) Ransac |
---|
| 57 | % 5) Bundle Adjustment |
---|
| 58 | % 6) up to scale translation reconstruction |
---|
| 59 | % 7) matches 3D triangulation |
---|
| 60 | % 8) Modified ImgInfo.Model.Depth up to accurate scale |
---|
| 61 | |
---|
| 62 | % initialize variables |
---|
| 63 | fail = 0; |
---|
| 64 | Pair.R = []; |
---|
| 65 | Pair.t = []; |
---|
| 66 | Pair.Xim = []; |
---|
| 67 | Pair.DepthScale = []; |
---|
| 68 | |
---|
| 69 | Img1 = ImgInfo(1).ExifInfo.IDName; |
---|
| 70 | Img2 = ImgInfo(2).ExifInfo.IDName; |
---|
| 71 | I1=imreadbw([defaultPara.Fdir '/pgm/' Img1 '.pgm']); % function from sift |
---|
| 72 | I2=imreadbw([defaultPara.Fdir '/pgm/' Img2 '.pgm']); % function from sift |
---|
| 73 | [f1] = readSurf(Img1, defaultPara.Fdir, 'Dense'); % Dense features |
---|
| 74 | [f2] = readSurf(Img2, defaultPara.Fdir, 'Dense'); % Dense features |
---|
| 75 | [D1 IND] = PorjPosi2Depth(size(I1), size(ImgInfo(1).Model.Depth.FitDepth), f1, ImgInfo(1).Model.Depth.FitDepth); |
---|
| 76 | [D2 IND] = PorjPosi2Depth(size(I2), size(ImgInfo(2).Model.Depth.FitDepth), f2, ImgInfo(1).Model.Depth.FitDepth); |
---|
| 77 | |
---|
| 78 | % 1) extract Measuesd Position and orientation from GPS or IMU info |
---|
| 79 | % Depends on what data we have, MeasR or MeasT, or both might be empty |
---|
| 80 | [MeasR MeasT] = InitPoseMeas(defaultPara, ImgInfo(1), ImgInfo(2)); |
---|
| 81 | |
---|
| 82 | if ~isempty(MeasR) |
---|
| 83 | % 2) Using Measures R and T and Mono-Depth to define match search space constrain |
---|
| 84 | % read in all surf features |
---|
| 85 | [ Rc1, Rc2, ConS1, ConS2, ConSRough1, ConSRough2] = CalMatchSearchRegin(defaultPara, MeasR, MeasT, I1, I2, f1, f2, D1, D2, 1, defaultPara.Flag.FlagDisp); |
---|
| 86 | |
---|
| 87 | % write the match search space constrain in to files for surfMatchRConS.sh script to read |
---|
| 88 | Vector2Ipoint([Rc1; ConS1],[defaultPara.Fdir '/surf/'],['RConS_' Img1]); |
---|
| 89 | Vector2Ipoint([Rc2; ConS2],[defaultPara.Fdir '/surf/'],['RConS_' Img2]); |
---|
| 90 | Vector2Ipoint([ConSRough1],[defaultPara.Fdir '/surf/'],['RConSRough_' Img1]); |
---|
| 91 | Vector2Ipoint([ConSRough2],[defaultPara.Fdir '/surf/'],['RConSRough_' Img2]); |
---|
| 92 | |
---|
| 93 | % 3) Do match search with all combinations satisfying Constrain from 2) using ralative threshould |
---|
| 94 | cd match |
---|
| 95 | [status, result] = system(['ls ' defaultPara.Fdir '/surf_matches/' Img1 '-' Img2 '.matchRConSDense_' num2str(defaultPara.AbsThre) '_' num2str(defaultPara.RatioThre)]); |
---|
| 96 | [statusReverse, resultReverse] ... |
---|
| 97 | = system(['ls ' defaultPara.Fdir '/surf_matches/' Img2 '-' Img1 '.matchRConSDense_' num2str(defaultPara.AbsThre) '_' num2str(defaultPara.RatioThre)]); |
---|
| 98 | if status && statusReverse |
---|
| 99 | SurfMatchTime = tic; |
---|
| 100 | system(['./surfMatchRConS.sh ' defaultPara.Fdir ' ' Img1 ' ' Img2 ' Dense ' num2str(defaultPara.AbsThre) ' ' num2str(defaultPara.RatioThre)]); |
---|
| 101 | disp([' ' num2str( toc( SurfMatchTime)) ' seconds.']); |
---|
| 102 | end |
---|
| 103 | cd .. |
---|
| 104 | else |
---|
| 105 | cd match |
---|
| 106 | [status, result] = system(['ls ' defaultPara.Fdir '/surf_matches/' Img1 '-' Img2 '.matchDense_' num2str(defaultPara.AbsThre) '_' num2str(defaultPara.RatioThre)]); |
---|
| 107 | [statusReverse, resultReverse] ... |
---|
| 108 | = system(['ls ' defaultPara.Fdir '/surf_matches/' Img2 '-' Img1 '.matchDense_' num2str(defaultPara.AbsThre) '_' num2str(defaultPara.RatioThre)]); |
---|
| 109 | if status && statusReverse |
---|
| 110 | SurfMatchTime = tic; |
---|
| 111 | system(['./surfMatch.sh ' defaultPara.Fdir ' ' Img1 ' ' Img2 ' Dense ' num2str(defaultPara.AbsThre) ' ' num2str(defaultPara.RatioThre)]); |
---|
| 112 | disp([' ' num2str( toc( SurfMatchTime)) ' seconds.']); |
---|
| 113 | end |
---|
| 114 | cd .. |
---|
| 115 | end |
---|
| 116 | % 4. Ransac |
---|
| 117 | if ~isempty(MeasR) |
---|
| 118 | [f1, f2, matches] = readSurfMatches(Img1, Img2, defaultPara.Fdir, ... |
---|
| 119 | [ defaultPara.Type 'Dense_' num2str(defaultPara.AbsThre) '_' num2str(defaultPara.RatioThre)], 1, 1); |
---|
| 120 | else |
---|
| 121 | [f1, f2, matches] = readSurfMatches(Img1, Img2, defaultPara.Fdir, ... |
---|
| 122 | [ 'Dense_' num2str(defaultPara.AbsThre) '_' num2str(defaultPara.RatioThre)], 1, 1); |
---|
| 123 | end |
---|
| 124 | if isempty(matches) |
---|
| 125 | disp('Zeros Surf matches'); |
---|
| 126 | fail = 1; |
---|
| 127 | return; |
---|
| 128 | end |
---|
| 129 | [D1 IND1] = PorjPosi2Depth(size(I1), size(ImgInfo(1).Model.Depth.FitDepth), f1(:,matches(1,:)), ImgInfo(1).Model.Depth.FitDepth); |
---|
| 130 | [D2 IND2] = PorjPosi2Depth(size(I2), size(ImgInfo(2).Model.Depth.FitDepth), f2(:,matches(2,:)), ImgInfo(1).Model.Depth.FitDepth); |
---|
| 131 | |
---|
| 132 | % Ensemble method to determine confidence of inliers |
---|
| 133 | fittingfn = @fundmatrix; |
---|
| 134 | distfnEnsmble = @fundistEnsmble; |
---|
| 135 | degenfn = @isdegenerate; |
---|
| 136 | nmatches = size(matches, 2); |
---|
| 137 | x = [[f1(:, matches(1,:)); ones(1, nmatches)]; [f2(:, matches(2,:)); ones(1, nmatches)]]; |
---|
| 138 | [ SampsonDist ] = EnsembleRansac(defaultPara, x, fittingfn, distfnEnsmble, degenfn, 8, ones(1,nmatches)', min(nmatches*10, defaultPara.MAXEnsembleSamples), 0); |
---|
| 139 | kurtosisValue =kurtosis(SampsonDist'); |
---|
| 140 | |
---|
| 141 | % Ransac |
---|
| 142 | [F0, inliers, NewDist, fail, ind]=GeneralRansac(defaultPara, f1, f2, matches, D1, D2, kurtosisValue', 8); |
---|
| 143 | if defaultPara.Flag.FlagDisp |
---|
| 144 | figure; plotmatches(I1,I2,f1, f2,matches(:,inliers), 'Stacking', 'v', 'Interactive', defaultPara.Flag.FlagDisp); |
---|
| 145 | end |
---|
| 146 | % *** Stop maunally to pick out the bad matches*** ----------------- |
---|
| 147 | matches = matches(:,inliers); |
---|
| 148 | if isempty(matches) |
---|
| 149 | disp('Zeros Matches After Ransac'); |
---|
| 150 | fail = 2; |
---|
| 151 | return; |
---|
| 152 | end |
---|
| 153 | % ------------------------------------------------------------------ |
---|
| 154 | |
---|
| 155 | x_calib = [ inv(defaultPara.InrinsicK1)*[ f1(:,matches(1,:)); ones(1,length(matches))];... |
---|
| 156 | inv(defaultPara.InrinsicK2)*[ f2(:,matches(2,:)); ones(1,length(matches))]]; |
---|
| 157 | |
---|
| 158 | % Estimate F using NonLine LS on every inlier |
---|
| 159 | MatchDensityWeights1 = CalMatchDensityWeights(f1(:,matches(1,:)), max(size(I1))/defaultPara.radius2imageSizeRatio); |
---|
| 160 | MatchDensityWeights2 = CalMatchDensityWeights(f2(:,matches(2,:)), max(size(I2))/defaultPara.radius2imageSizeRatio); |
---|
| 161 | MatchDensityWeights =mean([MatchDensityWeights1; MatchDensityWeights2], 1); |
---|
| 162 | F = getFnpt( F0, f1(:, matches(1,:))', f2(:, matches(2,:))', MatchDensityWeights); |
---|
| 163 | E = defaultPara.InrinsicK2'*F*defaultPara.InrinsicK1; % Camera essential Matrix |
---|
| 164 | if ~isempty(MeasR) |
---|
| 165 | [ R0, T0, lamda1, lamda2, inlier, Error] = EstPose( defaultPara, E, x_calib, [], MeasR(1:3,:)); |
---|
| 166 | else |
---|
| 167 | [ R0, T0, lamda1, lamda2, inlier, Error] = EstPose( defaultPara, E, x_calib, [], []); |
---|
| 168 | end |
---|
| 169 | T0 = [T0; - R0'*T0]; |
---|
| 170 | R0 = [R0; R0']; |
---|
| 171 | matches = matches(:,inlier); % delet matches give negative depths |
---|
| 172 | x_calib = x_calib(:,inlier); |
---|
| 173 | lamda1 = lamda1(inlier); |
---|
| 174 | lamda2 = lamda2(inlier); |
---|
| 175 | |
---|
| 176 | % Estimated X_obj by triangulation |
---|
| 177 | X_obj_1 = x_calib(1:3,:).*repmat(lamda1, 3, 1); |
---|
| 178 | X_obj_2 = R0(4:6,:)*(x_calib(4:6,:).*repmat(lamda2, 3, 1)) + repmat(T0(4:6), 1, size(matches,2)); |
---|
| 179 | X_obj = (X_obj_1+X_obj_2)/2; |
---|
| 180 | |
---|
| 181 | % 5. Bundle Adjustment |
---|
| 182 | [R T X_obj_BA X_im_BA dist1_BA dist2_BA]=SparseBAWraper(defaultPara, R0(1:3,:), T0(1:3), [f1(:,matches(1,:)); f2(:,matches(2,:))], X_obj, ImgInfo, 1); |
---|
| 183 | if all(isnan( dist1_BA)) || isempty(R) || any(isnan(R(:))) |
---|
| 184 | disp('BA failed'); |
---|
| 185 | fail = 3; |
---|
| 186 | return; |
---|
| 187 | end |
---|
| 188 | while length(X_im_BA) >= defaultPara.MinimumNumMatches |
---|
| 189 | outlier_thre1 = prctile(dist1_BA,90); |
---|
| 190 | outlier_thre2 = prctile(dist2_BA,90); |
---|
| 191 | Outlier = logical(zeros( size( dist1_BA))); |
---|
| 192 | if max(dist1_BA) >= defaultPara.ReProjErrorThre |
---|
| 193 | % Outlier = Outlier | dist1_BA > max( outlier_thre1, defaultPara.ReProjErrorThre); |
---|
| 194 | Outlier = Outlier | dist1_BA > outlier_thre1; |
---|
| 195 | end |
---|
| 196 | if max(dist2_BA) >= defaultPara.ReProjErrorThre |
---|
| 197 | % Outlier = Outlier | dist2_BA > max( outlier_thre2, defaultPara.ReProjErrorThre); |
---|
| 198 | Outlier = Outlier | dist2_BA > outlier_thre2; |
---|
| 199 | end |
---|
| 200 | matches(:,Outlier) = []; |
---|
| 201 | if isempty(matches) |
---|
| 202 | disp('Zeros Matches After BA Pruning'); |
---|
| 203 | fail = 4; |
---|
| 204 | return; |
---|
| 205 | end |
---|
| 206 | if all( Outlier == 0) |
---|
| 207 | % Non Outlier detected for BA |
---|
| 208 | break; |
---|
| 209 | end |
---|
| 210 | lamda1(Outlier) = []; |
---|
| 211 | lamda2(Outlier) = []; |
---|
| 212 | X_obj_BA(:,Outlier) = []; |
---|
| 213 | x_calib(:,Outlier) = []; |
---|
| 214 | [R T X_obj_BA X_im_BA dist1_BA dist2_BA]=SparseBAWraper(defaultPara, R, T, [f1(:,matches(1,:)); f2(:,matches(2,:))], X_obj_BA, ImgInfo, 1); |
---|
| 215 | if all(isnan( dist1_BA)) || isempty(R) || any(isnan(R(:))) |
---|
| 216 | disp('BA failed'); |
---|
| 217 | fail = 5; |
---|
| 218 | return; |
---|
| 219 | end |
---|
| 220 | end |
---|
| 221 | if defaultPara.Flag.FlagDisp |
---|
| 222 | figure; plotmatches(I1,I2,f1, f2, matches, 'Stacking', 'v', 'Interactive', defaultPara.Flag.FlagDisp); |
---|
| 223 | end |
---|
| 224 | |
---|
| 225 | % 6. find T up to scale |
---|
| 226 | |
---|
| 227 | % 7. Triangulation |
---|
| 228 | % modified the x_calib So that perfact triangulation but the image is distorted a little bit |
---|
| 229 | tempf1 = X_im_BA(1:2,:); |
---|
| 230 | tempf2 = X_im_BA(3:4,:); |
---|
| 231 | x_calib = [ inv(defaultPara.InrinsicK1)*[ tempf1; ones(1,length(tempf1))];... |
---|
| 232 | inv(defaultPara.InrinsicK2)*[ tempf2; ones(1,length(tempf2))]]; |
---|
| 233 | % ------------------ |
---|
| 234 | [ lamda1 lamda2 Error] = triangulation( defaultPara, R, T, x_calib); |
---|
| 235 | |
---|
| 236 | % 8. modify ImgInfo.Model.Depth .... (not sure do it or not??????) |
---|
| 237 | [D1 IND1] = PorjPosi2Depth(size(I1), size(ImgInfo(1).Model.Depth.FitDepth), f1(:,matches(1,:)), ImgInfo(1).Model.Depth.FitDepth); |
---|
| 238 | [D2 IND2] = PorjPosi2Depth(size(I2), size(ImgInfo(2).Model.Depth.FitDepth), f2(:,matches(2,:)), ImgInfo(2).Model.Depth.FitDepth); |
---|
| 239 | Depth1ProjDepthRatio = sqrt(sum(x_calib(1:3,:).^2, 1)); |
---|
| 240 | Depth2ProjDepthRatio = sqrt(sum(x_calib(4:6,:).^2, 1)); |
---|
| 241 | DProj1 = D1./Depth1ProjDepthRatio; |
---|
| 242 | DProj2 = D2./Depth2ProjDepthRatio; |
---|
| 243 | [DepthScale1] = UniformDepthScale( defaultPara, DProj1, lamda1, ones(1,length(lamda1))); |
---|
| 244 | [DepthScale2] = UniformDepthScale( defaultPara, DProj2, lamda2, ones(1,length(lamda2)) ); |
---|
| 245 | %if DepthScale1 > 20 | DepthScale1 <0.05 | DepthScale2 > 20 | DepthScale2 <0.05 %//Min used to use 10 and 0.2 |
---|
| 246 | % disp('Unrealistic in Rescaleing the depth, Check matchings'); |
---|
| 247 | % fail = -1; |
---|
| 248 | %end |
---|
| 249 | |
---|
| 250 | Pair.lamda = [lamda1; lamda2]; |
---|
| 251 | Pair.DepthScale = [DepthScale1; DepthScale2]; |
---|
| 252 | Pair.R = R; |
---|
| 253 | Pair.T = T; |
---|
| 254 | Pair.Xim = [f1(:,matches(1,:)); f2(:,matches(2,:))]; |
---|
| 255 | |
---|
| 256 | % check is triangulation reasonable |
---|
| 257 | if defaultPara.Flag.FlagDisp |
---|
| 258 | figure(50); clf; title('Closest point Match Point'); hold on; |
---|
| 259 | ClosestMatchPosition2 = x_calib(4:6,:).*repmat( lamda2, 3,1); |
---|
| 260 | ClosestMatchPosition1 = R*(x_calib(1:3,:).*repmat( lamda1, 3,1)) + repmat(T, 1, length(lamda1)); |
---|
| 261 | MonoStichPosition2 = x_calib(4:6,:).*repmat( DProj2.*DepthScale2, 3,1); |
---|
| 262 | MonoStichPosition1 = R*(x_calib(1:3,:).*repmat( DProj1.*DepthScale1, 3,1)) + repmat(T, 1, length(DProj1)); |
---|
| 263 | % ===================== |
---|
| 264 | [VDepth HDepth] = size(ImgInfo(2).Model.Depth.FitDepth); |
---|
| 265 | [VImg HImg] = size(I1); |
---|
| 266 | VIndexDepthRes = repmat((1:VDepth)', [1 HDepth]); |
---|
| 267 | HIndexDepthRes = repmat((1:HDepth), [VDepth 1]); |
---|
| 268 | VIndexImgRes = ( VIndexDepthRes -0.5)/VDepth*VImg; |
---|
| 269 | HIndexImgRes = ( HIndexDepthRes -0.5)/HDepth*HImg; |
---|
| 270 | ImgPositionPix = cat(3, HIndexImgRes, VIndexImgRes); |
---|
| 271 | All_x_calib = inv(defaultPara.InrinsicK1)*[ reshape( permute(ImgPositionPix, [ 3 1 2]), 2, []); ones(1, VDepth*HDepth)];% |
---|
| 272 | All_Ray = All_x_calib./repmat( sqrt(sum(All_x_calib.^2, 1)), 3, 1); |
---|
| 273 | All_Ray = repmat( All_Ray, 2, 1); |
---|
| 274 | % ==================== |
---|
| 275 | ReScaledPosi2 = All_Ray(4:6,:).*repmat( ImgInfo(2).Model.Depth.FitDepth(:)'*DepthScale2, 3,1); |
---|
| 276 | ReScaledPosi1 = R*(All_Ray(1:3,:).*repmat( ImgInfo(1).Model.Depth.FitDepth(:)'*DepthScale1, 3,1)) + repmat(T, 1, length(All_Ray)); |
---|
| 277 | ReScaledPosi2(:,IND2) = []; |
---|
| 278 | ReScaledPosi1(:,IND1) = []; |
---|
| 279 | scatter3(ReScaledPosi2(1,:)', ReScaledPosi2(3,:)', ReScaledPosi2(2,:)', 0.5*ones(1,size( ReScaledPosi2,2))); |
---|
| 280 | scatter3(ReScaledPosi1(1,:)', ReScaledPosi1(3,:)', ReScaledPosi1(2,:)', 1*ones(1,size( ReScaledPosi1,2))); |
---|
| 281 | scatter3(ClosestMatchPosition2(1,:)', ClosestMatchPosition2(3,:)', ClosestMatchPosition2(2,:)', 40, 'g'); |
---|
| 282 | scatter3(ClosestMatchPosition1(1,:)', ClosestMatchPosition1(3,:)', ClosestMatchPosition1(2,:)', 40, 'b'); |
---|
| 283 | line( [ ClosestMatchPosition2(1,:); ClosestMatchPosition1(1,:)], ... |
---|
| 284 | [ ClosestMatchPosition2(3,:); ClosestMatchPosition1(3,:)], ... |
---|
| 285 | [ ClosestMatchPosition2(2,:); ClosestMatchPosition1(2,:)]); |
---|
| 286 | % line( [ MonoStichPosition2(1,:); MonoStichPosition1(1,:)], ... |
---|
| 287 | % [ MonoStichPosition2(3,:); MonoStichPosition1(3,:)], ... |
---|
| 288 | % [ MonoStichPosition2(2,:); MonoStichPosition1(2,:)]); |
---|
| 289 | if ~isempty(ImgInfo(1).Model.Constrain.RayMatche) |
---|
| 290 | ClosestMatchPosition1Hist = R*(ImgInfo(1).Model.Constrain.RayMatche'.*repmat(ImgInfo(1).Model.Constrain.Depth_modified , 3, 1)) + repmat(T, 1, length(ImgInfo(1).Model.Constrain.RayMatche)); |
---|
| 291 | scatter3(ClosestMatchPosition1Hist(1,:)', ClosestMatchPosition1Hist(3,:)', ClosestMatchPosition1Hist(2,:)', 40, 'y'); |
---|
| 292 | end |
---|
| 293 | if ~isempty(ImgInfo(2).Model.Constrain.RayMatche) |
---|
| 294 | ClosestMatchPosition2Hist = ImgInfo(2).Model.Constrain.RayMatche'.*repmat(ImgInfo(2).Model.Constrain.Depth_modified , 3, 1); |
---|
| 295 | scatter3(ClosestMatchPosition2Hist(1,:)', ClosestMatchPosition2Hist(3,:)', ClosestMatchPosition2Hist(2,:)', 40, 'y'); |
---|
| 296 | end |
---|
| 297 | |
---|
| 298 | figure(51); clf; title('Closest point Match Point'); hold on; |
---|
| 299 | RawReScaledPosi2 = All_Ray(4:6,:).*repmat( ImgInfo(2).Model.Depth.RawDepth(:)'*DepthScale2, 3,1); |
---|
| 300 | RawReScaledPosi1 = R*(All_Ray(1:3,:).*repmat( ImgInfo(1).Model.Depth.RawDepth(:)'*DepthScale1, 3,1)) + repmat(T, 1, length(All_Ray)); |
---|
| 301 | RawReScaledPosi2(:,IND2) = []; |
---|
| 302 | RawReScaledPosi1(:,IND1) = []; |
---|
| 303 | scatter3(RawReScaledPosi2(1,:)', RawReScaledPosi2(3,:)', RawReScaledPosi2(2,:)', 1*ones(1,size( RawReScaledPosi2,2))); |
---|
| 304 | scatter3(RawReScaledPosi1(1,:)', RawReScaledPosi1(3,:)', RawReScaledPosi1(2,:)', 0.5*ones(1,size( RawReScaledPosi1,2))); |
---|
| 305 | scatter3(ClosestMatchPosition2(1,:)', ClosestMatchPosition2(3,:)', ClosestMatchPosition2(2,:)', 40, 'g'); |
---|
| 306 | scatter3(ClosestMatchPosition2(1,:)', ClosestMatchPosition2(3,:)', ClosestMatchPosition2(2,:)', 40, 'b'); |
---|
| 307 | line( [ ClosestMatchPosition2(1,:); ClosestMatchPosition1(1,:)], ... |
---|
| 308 | [ ClosestMatchPosition2(3,:); ClosestMatchPosition1(3,:)], ... |
---|
| 309 | [ ClosestMatchPosition2(2,:); ClosestMatchPosition1(2,:)]); |
---|
| 310 | line( [ MonoStichPosition2(1,:); MonoStichPosition1(1,:)], ... |
---|
| 311 | [ MonoStichPosition2(3,:); MonoStichPosition1(3,:)], ... |
---|
| 312 | [ MonoStichPosition2(2,:); MonoStichPosition1(2,:)]); |
---|
| 313 | end |
---|
| 314 | return; |
---|
| 315 | |
---|